├── DNN-NILM_Publication-List.xlsx
├── DNN-NILM_low-freq_Performance.xlsx
├── LICENSE
├── README.md
├── Visualize_F1.ipynb
├── Visualize_MAE.ipynb
└── main.bbl
/DNN-NILM_Publication-List.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ihomelab/dnn4nilm_overview/d17ca5dda3fc88784740991b9f5cc62cb6c0080b/DNN-NILM_Publication-List.xlsx
--------------------------------------------------------------------------------
/DNN-NILM_low-freq_Performance.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ihomelab/dnn4nilm_overview/d17ca5dda3fc88784740991b9f5cc62cb6c0080b/DNN-NILM_low-freq_Performance.xlsx
--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
1 | MIT License
2 |
3 | Copyright (c) 2019 ihomelab
4 |
5 | Permission is hereby granted, free of charge, to any person obtaining a copy
6 | of this software and associated documentation files (the "Software"), to deal
7 | in the Software without restriction, including without limitation the rights
8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 |
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 |
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
22 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # Review on Deep Neural Networks applied to Low-Frequency NILM
2 |
3 | This repo contains data and code that has been used for the publication
4 | "Review on Deep Neural Networks applied to Low-Frequency NILM" submitted @ MDPI
5 | Energies [doi.org/10.3390/en14092390](https://doi.org/10.3390/en14092390).
6 |
7 | This work is a considerable extension of the presentation "DNN for NILM on low
8 | frequency Data" that has been done at the NILM workshop 2019. You can find the
9 | corresponding presentation
10 | [here](https://www.youtube.com/watch?v=010fawyCOCs&list=PLJrF-gxa0ImryGeNtil-s9zPJOaV4w-Vy&index=11)
11 |
12 | Content:
13 | * `DNN-NILM_Publication-List.xlsx` contains the list of the DNN-NILM
14 | publications that have been reviewed in the mentioned publication. It
15 | corresponds with minor differences in columns and nomenclature to table 2 in
16 | the publication and is provided to allow for easy searching and filtering.
17 | Abbreviations are explained in the publication.
18 | * `Visualize_MAE.ipynb` and `Visualize_F1.ipynb` are the jupyter notebooks that
19 | have been used to generate the visualizations in the paper, i.e. figures 3
20 | and 4. Please be aware that citation numbering might have changed in the
21 | final publication.
22 | * `DNN-NILM_low-freq_Performance.xlsx` contains the list of metrics extracted
23 | from the reviewed publications. Publications that did
24 | * not report metrics,
25 | * report metrics other than F_1-score or MAE or
26 | * not report metrics according to the relevant evaluation scenario
27 | might not appear in the list. The file is the basis for the figures generated
28 | with the jupyter notebooks. Some explanations on the columns can be found in
29 | the tab `Explanations`. Please do not expect that *all* columns are filled up
30 | consistently.
31 |
32 | In case you are an author of one of the publications and feel that erroneous
33 | information has been compiled in our list, do either contact
34 | patrick.huber@hslu.ch or open a pull request with your suggested changes. We
35 | will appreciate your feedback!
36 |
37 |
--------------------------------------------------------------------------------
/Visualize_F1.ipynb:
--------------------------------------------------------------------------------
1 | {
2 | "cells": [
3 | {
4 | "cell_type": "code",
5 | "execution_count": 1,
6 | "metadata": {},
7 | "outputs": [],
8 | "source": [
9 | "import re\n",
10 | "\n",
11 | "import altair as alt\n",
12 | "from altair import datum\n",
13 | "import numpy as np\n",
14 | "import pandas as pd\n"
15 | ]
16 | },
17 | {
18 | "cell_type": "markdown",
19 | "metadata": {},
20 | "source": [
21 | "# Get Data\n",
22 | "## Extract Citation Numbers from Bibtex file"
23 | ]
24 | },
25 | {
26 | "cell_type": "code",
27 | "execution_count": 2,
28 | "metadata": {},
29 | "outputs": [],
30 | "source": [
31 | "p2bbl_file = r'main.bbl' # path to .bbl-file\n",
32 | "with open(p2bbl_file, 'r') as fp:\n",
33 | " fContent = fp.read()\n",
34 | " \n",
35 | "pattern = r'\\\\bibitem\\[.*?\\]\\{(.+?)\\}' # use raw string format 'r' -> see https://docs.python.org/3/howto/regex.html#the-backslash-plague\n",
36 | "re.compile(pattern)\n",
37 | "matchObj_list = re.findall(pattern, fContent, flags=re.DOTALL) # re.DOTALL is important because of newlines within brackets \n",
38 | "df_cit = pd.DataFrame({'Reference: Abbreviation': matchObj_list, 'Reference: Number': range(1,len(matchObj_list)+1)})\n",
39 | "df_cit['Reference: Number'] = '[' + df_cit['Reference: Number'].astype(str) + ']'\n",
40 | "df_cit['Reference: Abbreviation'] = df_cit['Reference: Abbreviation'].astype(str) \n",
41 | "# df_cit.head()"
42 | ]
43 | },
44 | {
45 | "cell_type": "code",
46 | "execution_count": 3,
47 | "metadata": {},
48 | "outputs": [],
49 | "source": [
50 | "## Read Performance Values"
51 | ]
52 | },
53 | {
54 | "cell_type": "code",
55 | "execution_count": 4,
56 | "metadata": {},
57 | "outputs": [],
58 | "source": [
59 | "path = r'DNN-NILM_low-freq_Performance.xlsx'\n",
60 | "df = pd.read_excel(path)\n",
61 | "unnamed_columns = []\n",
62 | "for column in df.columns:\n",
63 | " if column.find('Unnamed:') != -1:\n",
64 | " unnamed_columns.append(column)\n",
65 | "df = df.drop(columns=unnamed_columns)\n",
66 | "df['Reference: Abbreviation'] = df['Reference: Abbreviation'].astype(str) \n",
67 | "# df.head()"
68 | ]
69 | },
70 | {
71 | "cell_type": "code",
72 | "execution_count": 5,
73 | "metadata": {},
74 | "outputs": [],
75 | "source": [
76 | "# Sanitize data\n",
77 | "# replace '?' with -1 values in case of 'Input: Window Size'\n",
78 | "df.loc[ df['Input: Window Size'] == '?', 'Input: Window Size'] = -1\n",
79 | "df.loc[ df['Input: Window Size'] == 'nan', 'Input: Window Size' ] = -1\n",
80 | "df.loc[ df['Input: Window Size'] == 'not appl', 'Input: Window Size' ] = -1\n",
81 | "df.loc[ df['Input: Window Size'] == np.nan, 'Input: Window Size'] = -1\n",
82 | "df['Input: Window Size'] = df['Input: Window Size'].astype(float)"
83 | ]
84 | },
85 | {
86 | "cell_type": "code",
87 | "execution_count": 6,
88 | "metadata": {
89 | "jupyter": {
90 | "source_hidden": true
91 | }
92 | },
93 | "outputs": [],
94 | "source": [
95 | "# Join dataframes with citation keys\n",
96 | "df = df.merge(df_cit, on='Reference: Abbreviation', how='left')\n",
97 | "df = df.rename(columns={'Reference: Number_y': 'Reference: Number'})"
98 | ]
99 | },
100 | {
101 | "cell_type": "markdown",
102 | "metadata": {},
103 | "source": [
104 | "## Evaluation on Metrics for noised, unseen\n",
105 | "(Done directly in Excel):\n",
106 | "* classificaton\n",
107 | " * F1: 12 publications\n",
108 | "* regression\n",
109 | " * MAE: 20 publications\n",
110 | " * SAE: 12 publications\n",
111 | " * EstAcc: 7 publications"
112 | ]
113 | },
114 | {
115 | "cell_type": "markdown",
116 | "metadata": {},
117 | "source": [
118 | "# Investigate: F1 for noised, unseen case"
119 | ]
120 | },
121 | {
122 | "cell_type": "code",
123 | "execution_count": 7,
124 | "metadata": {},
125 | "outputs": [],
126 | "source": [
127 | "df_f1 = df[ (df['Performance Metric: Type'] == 'F1') & \n",
128 | " (df['Training: Type of Data'] == 'noised') & \n",
129 | " (df['Evaluation: Scenario'] == 'unseen')].copy()\n",
130 | "df_f1 = df_f1[['Reference: Abbreviation', 'Reference: Number', \n",
131 | " 'Appliance', 'Training: Dataset', 'Model: Basic Type', 'Model: Denomination', \n",
132 | " 'Input: Window Size', 'Input: Sampling Rate', 'Performance Metric: Value']].copy()"
133 | ]
134 | },
135 | {
136 | "cell_type": "code",
137 | "execution_count": 8,
138 | "metadata": {},
139 | "outputs": [
140 | {
141 | "name": "stdout",
142 | "output_type": "stream",
143 | "text": [
144 | "['kelly2015b' 'bonfigli2018' 'barsim2018' 'nascimento2016' 'murray2019'\n",
145 | " 'krystalakos2018' 'sudoso2019' 'linh2019' 'cavdar2019' 'yue2020'\n",
146 | " 'massidda2020' 'rafiq2020' 'kukunuri2020']\n",
147 | "13\n"
148 | ]
149 | }
150 | ],
151 | "source": [
152 | "print(df_f1['Reference: Abbreviation'].unique())\n",
153 | "print(len(df_f1['Reference: Abbreviation'].unique()))"
154 | ]
155 | },
156 | {
157 | "cell_type": "code",
158 | "execution_count": 9,
159 | "metadata": {},
160 | "outputs": [],
161 | "source": [
162 | "# Remove Baseline Model \n",
163 | "df_f1 = df_f1[ ~(df_f1['Model: Denomination'] == 'BL =0') ]\n",
164 | "# Note to self: Ke15, Zh2016 are baseline/reference models -> remove these Models from other authors\n",
165 | "df_f1 = df_f1[ ~(df_f1['Model: Denomination'].str.contains('Ke15') | \n",
166 | " df_f1['Model: Denomination'].str.contains('Zh2016') | \n",
167 | " df_f1['Model: Denomination'].str.contains('He2016')) ] "
168 | ]
169 | },
170 | {
171 | "cell_type": "code",
172 | "execution_count": 10,
173 | "metadata": {},
174 | "outputs": [
175 | {
176 | "name": "stdout",
177 | "output_type": "stream",
178 | "text": [
179 | "['CNN dAE' 'CNN-RNN' 'LSTM-bi' 'dAE' 'CNN RctAng' 'GRU-bi' 'CNN s2p'\n",
180 | " 'STL Iterative Pruning (30%)' 'STL Rank 4 Tensor D.' 'RNN' 'TP-NILM'\n",
181 | " 'RNN-GRU' 'CNN' 'RecCNN' 'MFS-LSTM' 'S2SwA' 'BERT']\n",
182 | "['dishwasher' 'fridge' 'kettle' 'microwave' 'washing machine'\n",
183 | " 'tumble dryer']\n",
184 | "['UK-DALE' 'REDD' 'REFIT']\n"
185 | ]
186 | }
187 | ],
188 | "source": [
189 | "# Group Results\n",
190 | "grouped = df_f1.groupby(['Reference: Abbreviation', 'Appliance', 'Training: Dataset'])\n",
191 | "list_of_maxs = []\n",
192 | "for group in grouped:\n",
193 | " group = group[1]\n",
194 | " min_row = group[ group['Performance Metric: Value'] == group['Performance Metric: Value'].max()]\n",
195 | " list_of_maxs.append(min_row)\n",
196 | "df_f1 = pd.concat(list_of_maxs)\n",
197 | "print(df_f1['Model: Denomination'].unique())\n",
198 | "print(df_f1['Appliance'].unique())\n",
199 | "print(df_f1['Training: Dataset'].unique())"
200 | ]
201 | },
202 | {
203 | "cell_type": "code",
204 | "execution_count": 11,
205 | "metadata": {},
206 | "outputs": [],
207 | "source": [
208 | "# Create column that combines 'Reference' and 'Number' to check that numbers are correct\n",
209 | "df_f1['No-Ref'] = df_f1['Reference: Abbreviation'] + '_' + df_f1['Reference: Number']"
210 | ]
211 | },
212 | {
213 | "cell_type": "code",
214 | "execution_count": 12,
215 | "metadata": {},
216 | "outputs": [
217 | {
218 | "name": "stdout",
219 | "output_type": "stream",
220 | "text": [
221 | "UK-DALE : 39\n",
222 | "REDD : 24\n",
223 | "REFIT : 5\n"
224 | ]
225 | }
226 | ],
227 | "source": [
228 | "datasets = df_f1['Training: Dataset'].unique()\n",
229 | "for dataset in datasets:\n",
230 | " print(dataset, ': ', np.sum(df_f1['Training: Dataset']==dataset))"
231 | ]
232 | },
233 | {
234 | "cell_type": "code",
235 | "execution_count": 13,
236 | "metadata": {},
237 | "outputs": [
238 | {
239 | "name": "stdout",
240 | "output_type": "stream",
241 | "text": [
242 | "dishwasher : 17\n",
243 | "fridge : 17\n",
244 | "kettle : 8\n",
245 | "microwave : 14\n",
246 | "washing machine : 11\n",
247 | "tumble dryer : 1\n"
248 | ]
249 | }
250 | ],
251 | "source": [
252 | "appliances = df_f1['Appliance'].unique()\n",
253 | "for app in appliances:\n",
254 | " print(app, ': ', np.sum(df_f1['Appliance'] == app))"
255 | ]
256 | },
257 | {
258 | "cell_type": "code",
259 | "execution_count": 14,
260 | "metadata": {},
261 | "outputs": [],
262 | "source": [
263 | "# remove appliances with low number of results\n",
264 | "df_f1 = df_f1[~(df_f1['Appliance'] == 'tumble dryer')]\n",
265 | "# # df_f1 = df_f1[~((df_f1['Appliance'] == 'lighting'))]\n",
266 | "# df_f1 = df_f1[~((df_f1['Appliance'] == 'cooker'))]\n",
267 | "# df_f1 = df_f1[~((df_f1['Appliance'] == 'shower'))]\n",
268 | "# df_f1 = df_f1[~((df_f1['Appliance'] == 'stove/oven'))]\n",
269 | "\n",
270 | "# remove Morgan2017 -> works on proprietary dataset & only fridge\n",
271 | "# df_f1 = df_f1[~(df_f1['Reference: Abbreviation'] == 'Morgan2017') ]"
272 | ]
273 | },
274 | {
275 | "cell_type": "code",
276 | "execution_count": 15,
277 | "metadata": {},
278 | "outputs": [
279 | {
280 | "name": "stdout",
281 | "output_type": "stream",
282 | "text": [
283 | "3.35\n"
284 | ]
285 | }
286 | ],
287 | "source": [
288 | "# Average number of results per appliance\n",
289 | "appliances = df_f1['Appliance'].unique()\n",
290 | "n_apps = 0\n",
291 | "for app in appliances:\n",
292 | " n_apps = n_apps + np.sum(df_f1['Appliance'] == app)\n",
293 | "print(n_apps/len(appliances)/4)"
294 | ]
295 | },
296 | {
297 | "cell_type": "code",
298 | "execution_count": 16,
299 | "metadata": {},
300 | "outputs": [
301 | {
302 | "data": {
303 | "text/plain": [
304 | "array([], dtype=object)"
305 | ]
306 | },
307 | "execution_count": 16,
308 | "metadata": {},
309 | "output_type": "execute_result"
310 | }
311 | ],
312 | "source": [
313 | "# find 'na' Reference: Numbers\n",
314 | "# df['Reference: Number'].isna()\n",
315 | "df['Reference: Abbreviation'][df['Reference: Number'].isna()].unique()"
316 | ]
317 | },
318 | {
319 | "cell_type": "markdown",
320 | "metadata": {},
321 | "source": [
322 | "## Overview "
323 | ]
324 | },
325 | {
326 | "cell_type": "code",
327 | "execution_count": 17,
328 | "metadata": {},
329 | "outputs": [
330 | {
331 | "data": {
332 | "text/html": [
333 | "\n",
334 | "
\n",
335 | ""
383 | ],
384 | "text/plain": [
385 | "alt.FacetChart(...)"
386 | ]
387 | },
388 | "execution_count": 17,
389 | "metadata": {},
390 | "output_type": "execute_result"
391 | }
392 | ],
393 | "source": [
394 | "# Visualization ,\n",
395 | "upper_bound = 1 #14\n",
396 | "sortByDataset = alt.Chart().mark_point(clip=True, size=90, filled=True).encode(\n",
397 | " x=alt.X('Training: Dataset:N', title='Dataset'),\n",
398 | " y=alt.Y('Performance Metric: Value:Q', title='F1-score', scale=alt.Scale(domain=(0, upper_bound))),\n",
399 | "# color=alt.Color('Reference: Abbreviation:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')),\n",
400 | " color=alt.Color('Reference: Number:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')),\n",
401 | "# color=alt.Color('Input: Sampling Rate:Q', legend=alt.Legend(title='Sampling Rate'), scale=alt.Scale(domain=(0,10), type='symlog')),\n",
402 | "# color=alt.Color('Input: Window Size:Q', legend=alt.Legend(title='Window Size'), scale=alt.Scale(type='log')),\n",
403 | ").transform_filter(\n",
404 | " alt.FieldOneOfPredicate(field='Training: Dataset', oneOf=['REDD', 'UK-DALE', 'REFIT'])\n",
405 | ")\n",
406 | "# text = alt.Chart().mark_text(\n",
407 | "# clip=True, align='left', dy=-5, dx=-24,\n",
408 | "# fontSize=13\n",
409 | "# ).encode(\n",
410 | "# x=alt.X('Training: Dataset:N'),\n",
411 | "# y=alt.Y('Performance Metric: Value:Q', title='MAE [W]', scale=alt.Scale(domain=(0, upper_bound))),\n",
412 | "# text = 'Reference: Abbreviation:N'\n",
413 | "# )\n",
414 | "\n",
415 | "alt.layer(\n",
416 | " sortByDataset, \n",
417 | " data=df_f1, width=90#, height=500\n",
418 | ").facet(\n",
419 | " column=alt.Column('Appliance:N', title=None)\n",
420 | ").configure_facet(\n",
421 | " spacing=7\n",
422 | ").configure_header(\n",
423 | " titleFontSize=14,\n",
424 | " labelFontSize=14\n",
425 | ")"
426 | ]
427 | },
428 | {
429 | "cell_type": "markdown",
430 | "metadata": {},
431 | "source": [
432 | "## Variant ZOOM"
433 | ]
434 | },
435 | {
436 | "cell_type": "code",
437 | "execution_count": 18,
438 | "metadata": {},
439 | "outputs": [],
440 | "source": [
441 | "# Individualized focus views\n",
442 | "b = 20 \n",
443 | "h = 250\n",
444 | "params = {\n",
445 | " 'Kettle': { 'app' : 'kettle', 'width': b*1, 'height': h, 'upper_bound': 1, 'lower_bound': 0.85 }, \n",
446 | " 'Microwave': { 'app' : 'microwave', 'width': b*3, 'height': h, 'upper_bound': 1, 'lower_bound': 0.85 }, \n",
447 | " 'Dishwasher': { 'app' : 'dishwasher', 'width': b*3, 'height': h, 'upper_bound': 1, 'lower_bound': 0.7 },\n",
448 | " 'Washing machine': { 'app' : 'washing machine', 'width': b*3, 'height': h, 'upper_bound': 1, 'lower_bound': 0.7 },\n",
449 | " 'Fridge': { 'app' : 'fridge', 'width': b*3, 'height': h, 'upper_bound': 1, 'lower_bound': 0.7 }, \n",
450 | "}"
451 | ]
452 | },
453 | {
454 | "cell_type": "code",
455 | "execution_count": 19,
456 | "metadata": {},
457 | "outputs": [
458 | {
459 | "data": {
460 | "text/html": [
461 | "\n",
462 | "\n",
463 | ""
511 | ],
512 | "text/plain": [
513 | "alt.HConcatChart(...)"
514 | ]
515 | },
516 | "execution_count": 19,
517 | "metadata": {},
518 | "output_type": "execute_result"
519 | }
520 | ],
521 | "source": [
522 | "# Visualization Dishwasher,\n",
523 | "layers = []\n",
524 | "for key, par in params.items():\n",
525 | " upper_bound = par['upper_bound']\n",
526 | " lower_bound = par['lower_bound']\n",
527 | " width = par['width']\n",
528 | " height = par['height']\n",
529 | " app = par['app']\n",
530 | " sortByDataset = alt.Chart().mark_point(clip=True, size=90, filled=True).encode(\n",
531 | " x=alt.X('Training: Dataset:N', title='Dataset'),\n",
532 | " y=alt.Y('Performance Metric: Value:Q', title='F₁', scale=alt.Scale(domain=(lower_bound, upper_bound))),\n",
533 | " color=alt.Color('Reference: Number:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')), \n",
534 | "# color=alt.Color('No-Ref:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')), \n",
535 | "# color=alt.Color('Reference: Abbreviation:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')), \n",
536 | "# shape=alt.Shape('Model: Basic Type', sort=['recurrent', 'feedforward'], legend=alt.Legend(title='Network Type'))\n",
537 | "# shape=alt.Shape('Model: Basic Type', sort=['recurrent', 'feedforward'], legend=alt.Legend(title='Network Type'))\n",
538 | " ).transform_filter(\n",
539 | " alt.FieldOneOfPredicate(field='Appliance', oneOf=[app])\n",
540 | " ).properties(\n",
541 | " title=key\n",
542 | " )\n",
543 | "# text = alt.Chart().mark_text(\n",
544 | "# clip=True, align='left', dy=-7, dx=-15,\n",
545 | "# fontSize=13\n",
546 | "# ).encode(\n",
547 | "# x=alt.X('Training: Dataset:N'),\n",
548 | "# y=alt.Y('Performance Metric: Value:Q', title='MAE [W]', scale=alt.Scale(domain=(lower_bound, upper_bound))),\n",
549 | "# # text = 'Reference: Abbreviation:N'\n",
550 | "# text = 'Reference: Number:N'\n",
551 | "# ).transform_filter(\n",
552 | "# alt.FieldOneOfPredicate(field='Appliance', oneOf=[key])\n",
553 | "# )\n",
554 | " layers.append(alt.layer(\n",
555 | " sortByDataset, \n",
556 | " data=df_f1, width=width, height=height\n",
557 | " ))\n",
558 | "alt.hconcat(*layers, spacing=50, bounds='flush')"
559 | ]
560 | },
561 | {
562 | "cell_type": "markdown",
563 | "metadata": {},
564 | "source": [
565 | "Note to self: I double checked the values for the top result in the corresponding publications. *They are correct.*"
566 | ]
567 | },
568 | {
569 | "cell_type": "code",
570 | "execution_count": 20,
571 | "metadata": {},
572 | "outputs": [
573 | {
574 | "data": {
575 | "text/html": [
576 | "\n",
577 | "\n",
578 | ""
626 | ],
627 | "text/plain": [
628 | "alt.FacetChart(...)"
629 | ]
630 | },
631 | "execution_count": 20,
632 | "metadata": {},
633 | "output_type": "execute_result"
634 | }
635 | ],
636 | "source": [
637 | "# VERSION FOR PUBLICATION\n",
638 | "# Visualization ,\n",
639 | "upper_bound = 1 #14\n",
640 | "lower_bound = 0.7\n",
641 | "sortByDataset = alt.Chart().mark_point(clip=True, size=90, filled=True).encode(\n",
642 | " x=alt.X('Training: Dataset:N', title='Dataset'),\n",
643 | " y=alt.Y('Performance Metric: Value:Q', title='F1-score', scale=alt.Scale(domain=(lower_bound, upper_bound))),\n",
644 | "# color=alt.Color('Reference: Abbreviation:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')),\n",
645 | " color=alt.Color('Reference: Number:N', legend=alt.Legend(title='Publication'), scale=alt.Scale(scheme = 'category20')),\n",
646 | ")\n",
647 | "\n",
648 | "alt.layer(\n",
649 | " sortByDataset, \n",
650 | " data=df_f1, width=90#, height=500\n",
651 | ").facet(\n",
652 | " column=alt.Column('Appliance:N', title=None)\n",
653 | ").configure_facet(\n",
654 | " spacing=7\n",
655 | ").configure_header(\n",
656 | " titleFontSize=14,\n",
657 | " labelFontSize=14\n",
658 | ")"
659 | ]
660 | },
661 | {
662 | "cell_type": "code",
663 | "execution_count": 21,
664 | "metadata": {},
665 | "outputs": [
666 | {
667 | "name": "stdout",
668 | "output_type": "stream",
669 | "text": [
670 | "kettle\n",
671 | "2\n",
672 | " Reference: Abbreviation Reference: Number Appliance Training: Dataset \\\n",
673 | "1497 rafiq2020 [64] kettle UK-DALE \n",
674 | "1048 sudoso2019 [109] kettle UK-DALE \n",
675 | "\n",
676 | " Model: Basic Type Model: Denomination Input: Window Size \\\n",
677 | "1497 combined MFS-LSTM -1.0 \n",
678 | "1048 combined S2SwA 768.0 \n",
679 | "\n",
680 | " Input: Sampling Rate Performance Metric: Value No-Ref \n",
681 | "1497 1 0.965 rafiq2020_[64] \n",
682 | "1048 6 0.967 sudoso2019_[109] \n",
683 | "\n",
684 | "microwave\n",
685 | "4\n",
686 | " Reference: Abbreviation Reference: Number Appliance Training: Dataset \\\n",
687 | "584 murray2019 [56] microwave REDD \n",
688 | "587 murray2019 [56] microwave REDD \n",
689 | "498 nascimento2016 [121] microwave REDD \n",
690 | "1199 cavdar2019 [144] microwave REDD \n",
691 | "\n",
692 | " Model: Basic Type Model: Denomination Input: Window Size \\\n",
693 | "584 feedforward CNN 720.0 \n",
694 | "587 recurrent RNN-GRU 720.0 \n",
695 | "498 recurrent RecCNN 764.0 \n",
696 | "1199 combined CNN-RNN -1.0 \n",
697 | "\n",
698 | " Input: Sampling Rate Performance Metric: Value No-Ref \n",
699 | "584 8 0.9500 murray2019_[56] \n",
700 | "587 8 0.9500 murray2019_[56] \n",
701 | "498 4 0.9537 nascimento2016_[121] \n",
702 | "1199 4 0.9578 cavdar2019_[144] \n",
703 | "\n",
704 | "dishwasher\n",
705 | "4\n",
706 | " Reference: Abbreviation Reference: Number Appliance Training: Dataset \\\n",
707 | "1392 massidda2020 [134] dishwasher UK-DALE \n",
708 | "603 murray2019 [56] dishwasher REFIT \n",
709 | "649 murray2019 [56] dishwasher REFIT \n",
710 | "515 nascimento2016 [121] dishwasher REDD \n",
711 | "\n",
712 | " Model: Basic Type Model: Denomination Input: Window Size \\\n",
713 | "1392 feedforward TP-NILM 30600.0 \n",
714 | "603 feedforward CNN 2400.0 \n",
715 | "649 recurrent RNN-GRU 2400.0 \n",
716 | "515 recurrent GRU-bi 2352.0 \n",
717 | "\n",
718 | " Input: Sampling Rate Performance Metric: Value No-Ref \n",
719 | "1392 60 0.8090 massidda2020_[134] \n",
720 | "603 8 0.8200 murray2019_[56] \n",
721 | "649 8 0.8200 murray2019_[56] \n",
722 | "515 4 0.8242 nascimento2016_[121] \n",
723 | "\n",
724 | "washing machine\n",
725 | "3\n",
726 | " Reference: Abbreviation Reference: Number Appliance \\\n",
727 | "1501 rafiq2020 [64] washing machine \n",
728 | "651 murray2019 [56] washing machine \n",
729 | "1393 massidda2020 [134] washing machine \n",
730 | "\n",
731 | " Training: Dataset Model: Basic Type Model: Denomination \\\n",
732 | "1501 UK-DALE combined MFS-LSTM \n",
733 | "651 REFIT recurrent RNN-GRU \n",
734 | "1393 UK-DALE feedforward TP-NILM \n",
735 | "\n",
736 | " Input: Window Size Input: Sampling Rate Performance Metric: Value \\\n",
737 | "1501 -1.0 1 0.765 \n",
738 | "651 2400.0 8 0.860 \n",
739 | "1393 30600.0 60 0.863 \n",
740 | "\n",
741 | " No-Ref \n",
742 | "1501 rafiq2020_[64] \n",
743 | "651 murray2019_[56] \n",
744 | "1393 massidda2020_[134] \n",
745 | "\n",
746 | "fridge\n",
747 | "4\n",
748 | " Reference: Abbreviation Reference: Number Appliance Training: Dataset \\\n",
749 | "151 barsim2018 [63] fridge UK-DALE \n",
750 | "604 murray2019 [56] fridge REFIT \n",
751 | "543 nascimento2016 [121] fridge REDD \n",
752 | "1217 cavdar2019 [144] fridge REDD \n",
753 | "\n",
754 | " Model: Basic Type Model: Denomination Input: Window Size \\\n",
755 | "151 feedforward CNN dAE 10800.0 \n",
756 | "604 feedforward CNN 6400.0 \n",
757 | "543 recurrent RecCNN 9604.0 \n",
758 | "1217 combined CNN-RNN -1.0 \n",
759 | "\n",
760 | " Input: Sampling Rate Performance Metric: Value No-Ref \n",
761 | "151 1 0.9270 barsim2018_[63] \n",
762 | "604 8 0.9300 murray2019_[56] \n",
763 | "543 4 0.9497 nascimento2016_[121] \n",
764 | "1217 4 0.9502 cavdar2019_[144] \n",
765 | "\n"
766 | ]
767 | }
768 | ],
769 | "source": [
770 | "# Evaluate x best performing Models\n",
771 | "apps = ['kettle', 'microwave', 'dishwasher', 'washing machine', 'fridge']\n",
772 | "authors = []\n",
773 | "for app in apps:\n",
774 | " df_app = df_f1[ df_f1['Appliance'] == app]\n",
775 | " df_app = df_app.sort_values('Performance Metric: Value')\n",
776 | " first_i_values = round(len(df_app)/4)\n",
777 | " df_app = df_app.iloc[-first_i_values:,:]\n",
778 | " print(app)\n",
779 | " print(first_i_values)\n",
780 | " authors.extend(df_app['Reference: Abbreviation'].values)\n",
781 | " print(df_app)\n",
782 | "# print('publications: ', df_app['Reference: Abbreviation'].values)\n",
783 | "# print('values', df_app['Performance Metric: Value'].values)\n",
784 | "# print('datasets: ', df_app['Training: Dataset'].unique())\n",
785 | "# print('no of ff: ', np.sum(df_app['Model: Basic Type'] == 'feedforward'), \n",
786 | "# 'no of rnn: ', np.sum(df_app['Model: Basic Type'] == 'recurrent'), \n",
787 | "# 'no of combined', np.sum(df_app['Model: Basic Type'] == 'combined') )\n",
788 | " print()"
789 | ]
790 | },
791 | {
792 | "cell_type": "code",
793 | "execution_count": 22,
794 | "metadata": {},
795 | "outputs": [
796 | {
797 | "data": {
798 | "text/html": [
799 | "\n",
800 | "\n",
813 | "
\n",
814 | " \n",
815 | " \n",
816 | " | \n",
817 | " authors | \n",
818 | " counts | \n",
819 | "
\n",
820 | " \n",
821 | " \n",
822 | " \n",
823 | " 2 | \n",
824 | " murray2019 | \n",
825 | " 6 | \n",
826 | "
\n",
827 | " \n",
828 | " 3 | \n",
829 | " nascimento2016 | \n",
830 | " 3 | \n",
831 | "
\n",
832 | " \n",
833 | " 0 | \n",
834 | " rafiq2020 | \n",
835 | " 2 | \n",
836 | "
\n",
837 | " \n",
838 | " 4 | \n",
839 | " cavdar2019 | \n",
840 | " 2 | \n",
841 | "
\n",
842 | " \n",
843 | " 5 | \n",
844 | " massidda2020 | \n",
845 | " 2 | \n",
846 | "
\n",
847 | " \n",
848 | " 1 | \n",
849 | " sudoso2019 | \n",
850 | " 1 | \n",
851 | "
\n",
852 | " \n",
853 | " 6 | \n",
854 | " barsim2018 | \n",
855 | " 1 | \n",
856 | "
\n",
857 | " \n",
858 | "
\n",
859 | "
"
860 | ],
861 | "text/plain": [
862 | " authors counts\n",
863 | "2 murray2019 6\n",
864 | "3 nascimento2016 3\n",
865 | "0 rafiq2020 2\n",
866 | "4 cavdar2019 2\n",
867 | "5 massidda2020 2\n",
868 | "1 sudoso2019 1\n",
869 | "6 barsim2018 1"
870 | ]
871 | },
872 | "execution_count": 22,
873 | "metadata": {},
874 | "output_type": "execute_result"
875 | }
876 | ],
877 | "source": [
878 | "authors = np.asarray(authors)\n",
879 | "counts = []\n",
880 | "for author in pd.Series(authors).unique():\n",
881 | " count = np.sum(authors == author)\n",
882 | " counts.append(count)\n",
883 | "counts = pd.DataFrame({'authors': pd.Series(authors).unique(), 'counts': counts})\n",
884 | "counts.sort_values('counts', ascending=False)"
885 | ]
886 | },
887 | {
888 | "cell_type": "code",
889 | "execution_count": null,
890 | "metadata": {},
891 | "outputs": [],
892 | "source": []
893 | }
894 | ],
895 | "metadata": {
896 | "kernel_info": {
897 | "name": "genpurp_p36"
898 | },
899 | "kernelspec": {
900 | "display_name": "Python 3",
901 | "language": "python",
902 | "name": "python3"
903 | },
904 | "language_info": {
905 | "codemirror_mode": {
906 | "name": "ipython",
907 | "version": 3
908 | },
909 | "file_extension": ".py",
910 | "mimetype": "text/x-python",
911 | "name": "python",
912 | "nbconvert_exporter": "python",
913 | "pygments_lexer": "ipython3",
914 | "version": "3.8.5"
915 | },
916 | "nteract": {
917 | "version": "0.15.0"
918 | }
919 | },
920 | "nbformat": 4,
921 | "nbformat_minor": 4
922 | }
923 |
--------------------------------------------------------------------------------
/main.bbl:
--------------------------------------------------------------------------------
1 | \begin{thebibliography}{-------}
2 | \providecommand{\natexlab}[1]{#1}
3 |
4 | \bibitem[Hart(1985)]{hart1985}
5 | Hart, G.W.
6 | \newblock Prototype Nonintrusive Appliance Load Monitor.
7 | \newblock Technical Report~2, {MIT Energy Laboratory and Electric Power
8 | Research Institute}, 1985.
9 |
10 | \bibitem[Hart(1992)]{hart1992}
11 | Hart, G.W.
12 | \newblock Nonintrusive Appliance Load Monitoring.
13 | \newblock {\em Proceedings of the IEEE} {\bf 1992}, {\em 80},~1870--1891.
14 | \newblock
15 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/5.192069}{\detokenize{10.1109/5.192069}}}.
16 |
17 | \bibitem[Zeifman and Roth(2011)]{zeifman2011}
18 | Zeifman, M.; Roth, K.
19 | \newblock Nonintrusive Appliance Load Monitoring: {{Review}} and Outlook.
20 | \newblock {\em IEEE Transactions on Consumer Electronics} {\bf 2011}, {\em
21 | 57},~76--84.
22 | \newblock
23 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TCE.2011.5735484}{\detokenize{10.1109/TCE.2011.5735484}}}.
24 |
25 | \bibitem[Alcal{\'a}(2016)]{alcala2016}
26 | Alcal{\'a}, J.
27 | \newblock Non-{{Intrusive Load Monitoring}} Techniques for {{Activity}} of
28 | {{Daily Living}} Recognition.
29 | \newblock PhD thesis, Universidad de Alcal\'a, {Madrid}, 2016.
30 |
31 | \bibitem[Salani \em{et~al.}(2020)Salani, Derboni, Rivola, Medici, Nespoli,
32 | Rosato, and Rizzoli]{salani2020}
33 | Salani, M.; Derboni, M.; Rivola, D.; Medici, V.; Nespoli, L.; Rosato, F.;
34 | Rizzoli, A.E.
35 | \newblock Non Intrusive Load Monitoring for Demand Side Management.
36 | \newblock {\em Energy Informatics} {\bf 2020}, {\em 3},~25.
37 | \newblock
38 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1186/s42162-020-00128-2}{\detokenize{10.1186/s42162-020-00128-2}}}.
39 |
40 | \bibitem[{\c C}imen \em{et~al.}(2020){\c C}imen, {\c C}etinkaya, Vasquez, and
41 | Guerrero]{cimen2020b}
42 | {\c C}imen, H.; {\c C}etinkaya, N.; Vasquez, J.C.; Guerrero, J.M.
43 | \newblock A {{Microgrid Energy Management System}} Based on {{Non}}-{{Intrusive
44 | Load Monitoring}} via {{Multitask Learning}}.
45 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2020}, pp. 1--1.
46 | \newblock
47 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2020.3027491}{\detokenize{10.1109/TSG.2020.3027491}}}.
48 |
49 | \bibitem[Gupta \em{et~al.}(2010)Gupta, Reynolds, and Patel]{gupta2010}
50 | Gupta, S.; Reynolds, M.S.; Patel, S.N.
51 | \newblock {{ElectriSense}}: {{Single}}-Point {{Sensing Using EMI}} for
52 | {{Electrical Event Detection}} and {{Classification}} in the {{Home}}.
53 | \newblock Proceedings of the 12th {{ACM International Conference}} on
54 | {{Ubiquitous Computing}}; {ACM}: {New York, NY, USA}, 2010; {{UbiComp}} '10,
55 | pp. 139--148.
56 | \newblock
57 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/1864349.1864375}{\detokenize{10.1145/1864349.1864375}}}.
58 |
59 | \bibitem[{Uribe-P{\'e}rez} \em{et~al.}(2016){Uribe-P{\'e}rez}, Hern{\'a}ndez,
60 | {de la Vega}, and Angulo]{uribe-perez2016}
61 | {Uribe-P{\'e}rez}, N.; Hern{\'a}ndez, L.; {de la Vega}, D.; Angulo, I.
62 | \newblock State of the {{Art}} and {{Trends Review}} of {{Smart Metering}} in
63 | {{Electricity Grids}}.
64 | \newblock {\em Applied Sciences} {\bf 2016}, {\em 6},~68.
65 | \newblock
66 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/app6030068}{\detokenize{10.3390/app6030068}}}.
67 |
68 | \bibitem[Kim \em{et~al.}(2011)Kim, Marwah, Arlitt, Lyon, and Han]{kim2011}
69 | Kim, H.; Marwah, M.; Arlitt, M.; Lyon, G.; Han, J.
70 | \newblock Unsupervised {{Disaggregation}} of {{Low Frequency Power
71 | Measurements}}. In {\em Proceedings of the 2011 {{SIAM International
72 | Conference}} on {{Data Mining}}}; Liu, B.; Liu, H.; Clifton, C.; Washio, T.;
73 | Kamath, C., Eds.; {Society for Industrial and Applied Mathematics}:
74 | {Philadelphia, PA}, 2011; pp. 747--758.
75 | \newblock
76 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1137/1.9781611972818.64}{\detokenize{10.1137/1.9781611972818.64}}}.
77 |
78 | \bibitem[Kolter and Jaakkola(2012)]{kolter2012}
79 | Kolter, J.Z.; Jaakkola, T.S.
80 | \newblock Approximate {{Inference}} in {{Additive Factorial HMMs}} with
81 | {{Application}} to {{Energy Disaggregation}}.
82 | \newblock {{AISTATS}}, 2012, Vol.~22, pp. 1472--1482.
83 |
84 | \bibitem[Parson \em{et~al.}(2014)Parson, Ghosh, Weal, and Rogers]{parson2014a}
85 | Parson, O.; Ghosh, S.; Weal, M.; Rogers, A.
86 | \newblock An Unsupervised Training Method for Non-Intrusive Appliance Load
87 | Monitoring.
88 | \newblock {\em Artificial Intelligence} {\bf 2014}, {\em 217},~1--19.
89 | \newblock
90 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.artint.2014.07.010}{\detokenize{10.1016/j.artint.2014.07.010}}}.
91 |
92 | \bibitem[Makonin \em{et~al.}(2016)Makonin, Popowich, Baji{\'c}, Gill, and
93 | Bartram]{makonin2016}
94 | Makonin, S.; Popowich, F.; Baji{\'c}, I.V.; Gill, B.; Bartram, L.
95 | \newblock Exploiting {{HMM Sparsity}} to {{Perform Online Real}}-{{Time
96 | Nonintrusive Load Monitoring}}.
97 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2016}, {\em
98 | 7},~2575--2585.
99 | \newblock
100 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2015.2494592}{\detokenize{10.1109/TSG.2015.2494592}}}.
101 |
102 | \bibitem[Tabatabaei \em{et~al.}(2017)Tabatabaei, Dick, and Xu]{tabatabaei2017}
103 | Tabatabaei, S.M.; Dick, S.; Xu, W.
104 | \newblock Toward {{Non}}-{{Intrusive Load Monitoring}} via {{Multi}}-{{Label
105 | Classification}}.
106 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2017}, {\em 8},~26--40.
107 | \newblock
108 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2016.2584581}{\detokenize{10.1109/TSG.2016.2584581}}}.
109 |
110 | \bibitem[Kelly and Knottenbelt(2015)]{kelly2015b}
111 | Kelly, J.; Knottenbelt, W.
112 | \newblock Neural {{NILM}}: {{Deep Neural Networks Applied}} to {{Energy
113 | Disaggregation}}.
114 | \newblock Proceedings of the 2nd {{ACM International Conference}} on
115 | {{Embedded Systems}} for {{Energy}}-{{Efficient Built Environments}};
116 | {Association for Computing Machinery}: {New York, NY, USA}, 2015;
117 | {{BuildSys}} '15, pp. 55--64.
118 | \newblock
119 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2821650.2821672}{\detokenize{10.1145/2821650.2821672}}}.
120 |
121 | \bibitem[Mauch and Yang(2015)]{mauch2015}
122 | Mauch, L.; Yang, B.
123 | \newblock A New Approach for Supervised Power Disaggregation by Using a Deep
124 | Recurrent {{LSTM}} Network.
125 | \newblock Signal and {{Information Processing}} ({{GlobalSIP}}), 2015 {{IEEE
126 | Global Conference}} On; {IEEE}: {Orlando, FL, USA}, 2015; pp. 63--67.
127 | \newblock
128 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/GlobalSIP.2015.7418157}{\detokenize{10.1109/GlobalSIP.2015.7418157}}}.
129 |
130 | \bibitem[Zoha \em{et~al.}(2012)Zoha, Gluhak, Imran, and Rajasegarar]{zoha2012}
131 | Zoha, A.; Gluhak, A.; Imran, M.A.; Rajasegarar, S.
132 | \newblock Non-Intrusive Load Monitoring Approaches for Disaggregated Energy
133 | Sensing: {{A}} Survey.
134 | \newblock {\em Sensors} {\bf 2012}, {\em 12},~16838--16866.
135 |
136 | \bibitem[Bonfigli \em{et~al.}(2015)Bonfigli, Squartini, Fagiani, and
137 | Piazza]{bonfigli2015}
138 | Bonfigli, R.; Squartini, S.; Fagiani, M.; Piazza, F.
139 | \newblock Unsupervised Algorithms for Non-Intrusive Load Monitoring: {{An}}
140 | up-to-Date Overview.
141 | \newblock {IEEE}, 2015, pp. 1175--1180.
142 | \newblock
143 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/EEEIC.2015.7165334}{\detokenize{10.1109/EEEIC.2015.7165334}}}.
144 |
145 | \bibitem[Pereira and Nunes(2018)]{pereira2018}
146 | Pereira, L.; Nunes, N.
147 | \newblock Performance Evaluation in Non-Intrusive Load Monitoring:
148 | {{Datasets}}, Metrics, and Tools\textemdash{{A}} Review.
149 | \newblock {\em Wiley Interdisciplinary Reviews: Data Mining and Knowledge
150 | Discovery} {\bf 2018}, {\em 8},~e1265.
151 | \newblock
152 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1002/widm.1265}{\detokenize{10.1002/widm.1265}}}.
153 |
154 | \bibitem[Nalmpantis and Vrakas(2019)]{nalmpantis2019}
155 | Nalmpantis, C.; Vrakas, D.
156 | \newblock Machine Learning Approaches for Non-Intrusive Load Monitoring: From
157 | Qualitative to Quantitative Comparation.
158 | \newblock {\em Artificial Intelligence Review} {\bf 2019}, {\em 52},~217--243.
159 | \newblock
160 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/s10462-018-9613-7}{\detokenize{10.1007/s10462-018-9613-7}}}.
161 |
162 | \bibitem[Bonfigli and Squartini(2020)]{bonfigli2020}
163 | Bonfigli, R.; Squartini, S.
164 | \newblock {\em Machine {{Learning Approaches}} to {{Non}}-{{Intrusive Load
165 | Monitoring}}}; {{SpringerBriefs}} in {{Energy}}, {Springer International
166 | Publishing}: {Cham}, 2020.
167 | \newblock
168 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/978-3-030-30782-0}{\detokenize{10.1007/978-3-030-30782-0}}}.
169 |
170 | \bibitem[Bonfigli \em{et~al.}(2018)Bonfigli, Felicetti, Principi, Fagiani,
171 | Squartini, and Piazza]{bonfigli2018}
172 | Bonfigli, R.; Felicetti, A.; Principi, E.; Fagiani, M.; Squartini, S.; Piazza,
173 | F.
174 | \newblock Denoising Autoencoders for {{Non}}-{{Intrusive Load Monitoring}}:
175 | {{Improvements}} and Comparative Evaluation.
176 | \newblock {\em Energy and Buildings} {\bf 2018}, {\em 158},~1461--1474.
177 | \newblock
178 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.enbuild.2017.11.054}{\detokenize{10.1016/j.enbuild.2017.11.054}}}.
179 |
180 | \bibitem[Valenti \em{et~al.}(2018)Valenti, Bonfigli, Principi, and
181 | Squartini]{valenti2018}
182 | Valenti, M.; Bonfigli, R.; Principi, E.; Squartini, a.S.
183 | \newblock Exploiting the {{Reactive Power}} in {{Deep Neural Models}} for
184 | {{Non}}-{{Intrusive Load Monitoring}}.
185 | \newblock 2018 {{International Joint Conference}} on {{Neural Networks}}
186 | ({{IJCNN}}); {IEEE}: {Rio de Janeiro}, 2018; pp. 1--8.
187 | \newblock
188 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/IJCNN.2018.8489271}{\detokenize{10.1109/IJCNN.2018.8489271}}}.
189 |
190 | \bibitem[Batra \em{et~al.}(2019)Batra, Kukunuri, Pandey, Malakar, Kumar,
191 | Krystalakos, Zhong, Meira, and Parson]{batra2019a}
192 | Batra, N.; Kukunuri, R.; Pandey, A.; Malakar, R.; Kumar, R.; Krystalakos, O.;
193 | Zhong, M.; Meira, P.; Parson, O.
194 | \newblock Towards {{Reproducible State}}-of-the-Art {{Energy Disaggregation}}.
195 | \newblock Proceedings of the 6th {{ACM International Conference}} on
196 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
197 | {{Transportation}}; {ACM}: {New York, NY, USA}, 2019; {{BuildSys}} '19, pp.
198 | 193--202.
199 | \newblock
200 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3360322.3360844}{\detokenize{10.1145/3360322.3360844}}}.
201 |
202 | \bibitem[Reinhardt and Klemenjak(2020)]{reinhardt2020}
203 | Reinhardt, A.; Klemenjak, C.
204 | \newblock How Does {{Load Disaggregation Performance Depend}} on {{Data
205 | Characteristics}}? {{Insights}} from a {{Benchmarking Study}}.
206 | \newblock Proceedings of the {{Eleventh ACM International Conference}} on
207 | {{Future Energy Systems}}; {Association for Computing Machinery}: {Virtual
208 | Event, Australia}, 2020; E-{{Energy}} '20, pp. 167--177.
209 | \newblock
210 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3396851.3397691}{\detokenize{10.1145/3396851.3397691}}}.
211 |
212 | \bibitem[Batra \em{et~al.}(2014)Batra, Kelly, Parson, Dutta, Knottenbelt,
213 | Rogers, Singh, and Srivastava]{batra2014a}
214 | Batra, N.; Kelly, J.; Parson, O.; Dutta, H.; Knottenbelt, W.; Rogers, A.;
215 | Singh, A.; Srivastava, M.
216 | \newblock {{NILMTK}}: An Open Source Toolkit for Non-Intrusive Load Monitoring.
217 | \newblock {ACM Press}, 2014, pp. 265--276.
218 | \newblock
219 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2602044.2602051}{\detokenize{10.1145/2602044.2602051}}}.
220 |
221 | \bibitem[Kelly \em{et~al.}(2014)Kelly, Batra, Parson, Dutta, Knottenbelt,
222 | Rogers, Singh, and Srivastava]{kelly2014}
223 | Kelly, J.; Batra, N.; Parson, O.; Dutta, H.; Knottenbelt, W.; Rogers, A.;
224 | Singh, A.; Srivastava, M.
225 | \newblock {{NILMTK}} v0.2: A Non-Intrusive Load Monitoring Toolkit for Large
226 | Scale Data Sets: Demo Abstract.
227 | \newblock Proceedings of the 1st {{ACM Conference}} on {{Embedded Systems}}
228 | for {{Energy}}-{{Efficient Buildings}}; {ACM Press}: {Memphis, USA}, 2014;
229 | pp. 182--183.
230 | \newblock
231 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2674061.2675024}{\detokenize{10.1145/2674061.2675024}}}.
232 |
233 | \bibitem[Roos \em{et~al.}(1994)Roos, Lane, Botha, and Hancke]{roos1994}
234 | Roos, J.G.; Lane, I.E.; Botha, E.C.; Hancke, G.P.
235 | \newblock Using Neural Networks for Non-Intrusive Monitoring of Industrial
236 | Electrical Loads.
237 | \newblock Conference {{Proceedings}}. 10th {{Anniversary}}. {{IMTC}}/94.
238 | {{Advanced Technologies}} in {{I M}}. 1994 {{IEEE Instrumentation}} and
239 | {{Measurement Technolgy Conference}} ({{Cat}}. {{No}}.{{94CH3424}}-9), 1994,
240 | pp. 1115--1118 vol.3.
241 | \newblock
242 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/IMTC.1994.351862}{\detokenize{10.1109/IMTC.1994.351862}}}.
243 |
244 | \bibitem[Paradiso \em{et~al.}(2013)Paradiso, Paganelli, Luchetta, Giuli, and
245 | Castrogiovanni]{paradiso2013}
246 | Paradiso, F.; Paganelli, F.; Luchetta, A.; Giuli, D.; Castrogiovanni, P.
247 | \newblock {{ANN}}-Based Appliance Recognition from Low-Frequency Energy
248 | Monitoring Data.
249 | \newblock 2013 {{IEEE}} 14th {{International Symposium}} on "{{A World}} of
250 | {{Wireless}}, {{Mobile}} and {{Multimedia Networks}}" ({{WoWMoM}}), 2013,
251 | pp. 1--6.
252 | \newblock
253 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/WoWMoM.2013.6583496}{\detokenize{10.1109/WoWMoM.2013.6583496}}}.
254 |
255 | \bibitem[Li and Dick(2016)]{li2016}
256 | Li, D.; Dick, S.
257 | \newblock Whole-House {{Non}}-{{Intrusive Appliance Load Monitoring}} via
258 | Multi-Label Classification.
259 | \newblock 2016 {{International Joint Conference}} on {{Neural Networks}}
260 | ({{IJCNN}}), 2016, pp. 2749--2755.
261 | \newblock
262 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/IJCNN.2016.7727545}{\detokenize{10.1109/IJCNN.2016.7727545}}}.
263 |
264 | \bibitem[Salerno and Rabbeni(2018)]{salerno2018}
265 | Salerno, V.M.; Rabbeni, G.
266 | \newblock An {{Extreme Learning Machine Approach}} to {{Effective Energy
267 | Disaggregation}}.
268 | \newblock {\em Electronics} {\bf 2018}, {\em 7},~235.
269 | \newblock
270 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/electronics7100235}{\detokenize{10.3390/electronics7100235}}}.
271 |
272 | \bibitem[Verma \em{et~al.}(2019)Verma, Singh, and Majumdar]{verma2019a}
273 | Verma, S.; Singh, S.; Majumdar, A.
274 | \newblock Multi {{Label Restricted Boltzmann Machine}} for {{Non}}-Intrusive
275 | {{Load Monitoring}}.
276 | \newblock {{ICASSP}} 2019-2019 {{IEEE International Conference}} on
277 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}). {IEEE},
278 | 2019, pp. 8345--8349.
279 |
280 | \bibitem[Goodfellow \em{et~al.}(2016)Goodfellow, Bengio, and
281 | Courville]{goodfellow2016}
282 | Goodfellow, I.; Bengio, Y.; Courville, A.
283 | \newblock {\em Deep {{Learning}}}; {MIT Press}, 2016.
284 |
285 | \bibitem[Chollet(2018)]{chollet2018}
286 | Chollet, F.
287 | \newblock {\em Deep Learning with {{Python}}}; {Manning Publications Co}:
288 | {Shelter Island, New York}, 2018.
289 |
290 | \bibitem[Chollet(2015)]{chollet2015}
291 | Chollet, F.
292 | \newblock Keras: The {{Python}} Deep Learning {{API}}, 2015.
293 |
294 | \bibitem[Shin \em{et~al.}(2019)Shin, Lee, Han, Yim, Rhee, and Lee]{shin2019}
295 | Shin, C.; Lee, E.; Han, J.; Yim, J.; Rhee, W.; Lee, H.
296 | \newblock The {{ENERTALK}} Dataset, 15 {{Hz}} Electricity Consumption Data from
297 | 22 Houses in {{Korea}}.
298 | \newblock {\em Scientific Data} {\bf 2019}, {\em 6},~193.
299 | \newblock
300 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/s41597-019-0212-5}{\detokenize{10.1038/s41597-019-0212-5}}}.
301 |
302 | \bibitem[Himeur \em{et~al.}(2020)Himeur, Alsalemi, Bensaali, and
303 | Amira]{himeur2020}
304 | Himeur, Y.; Alsalemi, A.; Bensaali, F.; Amira, A.
305 | \newblock Building Power Consumption Datasets: {{Survey}}, Taxonomy and Future
306 | Directions.
307 | \newblock {\em Energy and Buildings} {\bf 2020}, {\em 227},~110404.
308 | \newblock
309 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.enbuild.2020.110404}{\detokenize{10.1016/j.enbuild.2020.110404}}}.
310 |
311 | \bibitem[Huber \em{et~al.}(2020)Huber, Ott, Friedli, Rumsch, and
312 | Paice]{huber2020}
313 | Huber, P.; Ott, M.; Friedli, M.; Rumsch, A.; Paice, A.
314 | \newblock Residential {{Power Traces}} for {{Five Houses}}: {{The iHomeLab RAPT
315 | Dataset}}.
316 | \newblock {\em Data} {\bf 2020}, {\em 5},~17.
317 | \newblock
318 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/data5010017}{\detokenize{10.3390/data5010017}}}.
319 |
320 | \bibitem[Klemenjak \em{et~al.}(2020)Klemenjak, Kovatsch, Herold, and
321 | Elmenreich]{klemenjak2020}
322 | Klemenjak, C.; Kovatsch, C.; Herold, M.; Elmenreich, W.
323 | \newblock A Synthetic Energy Dataset for Non-Intrusive Load Monitoring in
324 | Households.
325 | \newblock {\em Scientific Data} {\bf 2020}, {\em 7},~108.
326 | \newblock
327 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/s41597-020-0434-6}{\detokenize{10.1038/s41597-020-0434-6}}}.
328 |
329 | \bibitem[V{\"o}lker \em{et~al.}(2020)V{\"o}lker, Pfeifer, Scholl, and
330 | Becker]{volker2020}
331 | V{\"o}lker, B.; Pfeifer, M.; Scholl, P.M.; Becker, B.
332 | \newblock {{FIRED}}: {{A Fully}}-Labeled {{hIgh}}-{{fRequency Electricity
333 | Disaggregation Dataset}}.
334 | \newblock Proceedings of the 7th {{ACM International Conference}} on
335 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
336 | {{Transportation}}; {Association for Computing Machinery}: {New York, NY,
337 | USA}, 2020; {{BuildSys}} '20, pp. 294--297.
338 | \newblock
339 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3408308.3427623}{\detokenize{10.1145/3408308.3427623}}}.
340 |
341 | \bibitem[Brewitt and Goddard(2018)]{brewitt2018}
342 | Brewitt, C.; Goddard, N.
343 | \newblock Non-{{Intrusive Load Monitoring}} with {{Fully Convolutional
344 | Networks}}.
345 | \newblock {\em arXiv:1812.03915 [cs, stat]} {\bf 2018},
346 | \href{http://xxx.lanl.gov/abs/1812.03915}{{\normalfont [arXiv:cs,
347 | stat/1812.03915]}}.
348 |
349 | \bibitem[Kelly and Knottenbelt(2015)]{kelly2015a}
350 | Kelly, J.; Knottenbelt, W.
351 | \newblock The {{UK}}-{{DALE}} Dataset, Domestic Appliance-Level Electricity
352 | Demand and Whole-House Demand from Five {{UK}} Homes.
353 | \newblock {\em Scientific Data} {\bf 2015}, {\em 2},~150007.
354 | \newblock
355 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/sdata.2015.7}{\detokenize{10.1038/sdata.2015.7}}}.
356 |
357 | \bibitem[Kolter and Johnson(2011)]{kolter2011}
358 | Kolter, J.Z.; Johnson, M.J.
359 | \newblock {{REDD}}: {{A}} Public Data Set for Energy Disaggregation Research.
360 | \newblock Workshop on {{Data Mining Applications}} in {{Sustainability}}
361 | ({{SIGKDD}}), {{San Diego}}, {{CA}}. {Citeseer}, 2011, Vol.~25, pp. 59--62.
362 |
363 | \bibitem[Makonin \em{et~al.}(2013)Makonin, Popowich, Bartram, Gill, and
364 | Bajic]{makonin2013}
365 | Makonin, S.; Popowich, F.; Bartram, L.; Gill, B.; Bajic, I.
366 | \newblock {{AMPds}}: {{A}} Public Dataset for Load Disaggregation and
367 | Eco-Feedback Research.
368 | \newblock 2013 {{IEEE Electrical Power Energy Conference}} ({{EPEC}}), 2013,
369 | pp. 1--6.
370 | \newblock
371 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/EPEC.2013.6802949}{\detokenize{10.1109/EPEC.2013.6802949}}}.
372 |
373 | \bibitem[Makonin(2015)]{makonin2015a}
374 | Makonin, S.
375 | \newblock {{AMPds}}: {{Almanac}} of {{Minutely Power}} Dataset ({{R2013}}),
376 | 2015.
377 | \newblock
378 | doi:{\changeurlcolor{black}\href{https://doi.org/10.7910/DVN/MXB7VO}{\detokenize{10.7910/DVN/MXB7VO}}}.
379 |
380 | \bibitem[Makonin \em{et~al.}(2016)Makonin, Ellert, Baji{\'c}, and
381 | Popowich]{makonin2016a}
382 | Makonin, S.; Ellert, B.; Baji{\'c}, I.V.; Popowich, F.
383 | \newblock Electricity, Water, and Natural Gas Consumption of a Residential
384 | House in {{Canada}} from 2012 to 2014.
385 | \newblock {\em Scientific Data} {\bf 2016}, {\em 3},~160037.
386 | \newblock
387 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/sdata.2016.37}{\detokenize{10.1038/sdata.2016.37}}}.
388 |
389 | \bibitem[Makonin(2016)]{makonin2016b}
390 | Makonin, S.
391 | \newblock {{AMPds2}}: {{The Almanac}} of {{Minutely Power}} Dataset
392 | ({{Version}} 2), 2016.
393 | \newblock
394 | doi:{\changeurlcolor{black}\href{https://doi.org/10.7910/DVN/FIE0S4}{\detokenize{10.7910/DVN/FIE0S4}}}.
395 |
396 | \bibitem[Murray \em{et~al.}(2017)Murray, Stankovic, and Stankovic]{murray2017}
397 | Murray, D.; Stankovic, L.; Stankovic, V.
398 | \newblock An Electrical Load Measurements Dataset of {{United Kingdom}}
399 | Households from a Two-Year Longitudinal Study.
400 | \newblock {\em Scientific data} {\bf 2017}, {\em 4},~160122.
401 | \newblock
402 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/sdata.2016.122}{\detokenize{10.1038/sdata.2016.122}}}.
403 |
404 | \bibitem[Parson \em{et~al.}(2015)Parson, Fisher, Hersey, Batra, Kelly, Singh,
405 | Knottenbelt, and Rogers]{parson2015}
406 | Parson, O.; Fisher, G.; Hersey, A.; Batra, N.; Kelly, J.; Singh, A.;
407 | Knottenbelt, W.; Rogers, A.
408 | \newblock Dataport and {{NILMTK}}: {{A}} Building Data Set Designed for
409 | Non-Intrusive Load Monitoring.
410 | \newblock 2015 {{IEEE Global Conference}} on {{Signal}} and {{Information
411 | Processing}} ({{GlobalSIP}}), 2015, pp. 210--214.
412 | \newblock
413 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/GlobalSIP.2015.7418187}{\detokenize{10.1109/GlobalSIP.2015.7418187}}}.
414 |
415 | \bibitem[Beckel \em{et~al.}(2014)Beckel, Kleiminger, Cicchetti, Staake, and
416 | Santini]{beckel2014}
417 | Beckel, C.; Kleiminger, W.; Cicchetti, R.; Staake, T.; Santini, S.
418 | \newblock The {{ECO}} Data Set and the Performance of Non-Intrusive Load
419 | Monitoring Algorithms.
420 | \newblock Proceedings of the 1st {{ACM Conference}} on {{Embedded Systems}}
421 | for {{Energy}}-{{Efficient Buildings}}; {Association for Computing
422 | Machinery}: {New York, NY, USA}, 2014; {{BuildSys}} '14, pp. 80--89.
423 | \newblock
424 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2674061.2674064}{\detokenize{10.1145/2674061.2674064}}}.
425 |
426 | \bibitem[Uttama~Nambi \em{et~al.}(2015)Uttama~Nambi, Reyes~Lua, and
427 | Prasad]{uttamanambi2015}
428 | Uttama~Nambi, A.S.; Reyes~Lua, A.; Prasad, V.R.
429 | \newblock {{LocED}}: {{Location}}-Aware {{Energy Disaggregation Framework}}.
430 | \newblock Proceedings of the 2nd {{ACM International Conference}} on
431 | {{Embedded Systems}} for {{Energy}}-{{Efficient Built Environments}};
432 | {Association for Computing Machinery}: {New York, NY, USA}, 2015;
433 | {{BuildSys}} '15, pp. 45--54.
434 | \newblock
435 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2821650.2821659}{\detokenize{10.1145/2821650.2821659}}}.
436 |
437 | \bibitem[Zimmermann \em{et~al.}(2012)Zimmermann, Evans, Griggs, King, Harding,
438 | Roberts, and Evans]{zimmermann2012}
439 | Zimmermann, J.P.; Evans, M.; Griggs, J.; King, N.; Harding, L.; Roberts, P.;
440 | Evans, C.
441 | \newblock Household {{Electricity Survey}}: {{A}} Study of Domestic Electrical
442 | Product Usage.
443 | \newblock {\em Intertek Testing \& Certification Ltd} {\bf 2012}, {\em Intertek
444 | Report R66141},~600.
445 |
446 | \bibitem[Martins \em{et~al.}(2018)Martins, Nascimento, {de Freitas},
447 | {Bittencourt e Silva}, and Guimar{\~a}es Duarte~Pinto]{martins2018a}
448 | Martins, P.B.M.; Nascimento, V.B.; {de Freitas}, A.R.; {Bittencourt e Silva},
449 | P.; Guimar{\~a}es Duarte~Pinto, R.
450 | \newblock Industrial {{Machines Dataset}} for {{Electrical Load
451 | Disaggregation}}, 2018.
452 | \newblock
453 | doi:{\changeurlcolor{black}\href{https://doi.org/10.21227/cg5v-dk02}{\detokenize{10.21227/cg5v-dk02}}}.
454 |
455 | \bibitem[Gao \em{et~al.}(2014)Gao, Giri, Kara, and Berg{\'e}s]{gao2014}
456 | Gao, J.; Giri, S.; Kara, E.C.; Berg{\'e}s, M.
457 | \newblock {{PLAID}}: A Public Dataset of High-Resoultion Electrical Appliance
458 | Measurements for Load Identification Research: Demo Abstract.
459 | \newblock Proceedings of the 1st {{ACM Conference}} on {{Embedded Systems}}
460 | for {{Energy}}-{{Efficient Buildings}}; {Association for Computing
461 | Machinery}: {New York, NY, USA}, 2014; {{BuildSys}} '14, pp. 198--199.
462 | \newblock
463 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2674061.2675032}{\detokenize{10.1145/2674061.2675032}}}.
464 |
465 | \bibitem[Klemenjak \em{et~al.}(2019)Klemenjak, Faustine, Makonin, and
466 | Elmenreich]{klemenjak2019a}
467 | Klemenjak, C.; Faustine, A.; Makonin, S.; Elmenreich, W.
468 | \newblock On {{Metrics}} to {{Assess}} the {{Transferability}} of {{Machine
469 | Learning Models}} in {{Non}}-{{Intrusive Load Monitoring}}.
470 | \newblock {\em arXiv preprint arXiv:1912.06200} {\bf 2019},
471 | \href{http://xxx.lanl.gov/abs/1912.06200}{{\normalfont [1912.06200]}}.
472 |
473 | \bibitem[DIncecco \em{et~al.}(2019)DIncecco, Squartini, and
474 | Zhong]{dincecco2019}
475 | DIncecco, M.; Squartini, S.; Zhong, M.
476 | \newblock Transfer {{Learning}} for {{Non}}-{{Intrusive Load Monitoring}}.
477 | \newblock {\em arXiv:1902.08835 [cs, stat]} {\bf 2019},
478 | \href{http://xxx.lanl.gov/abs/1902.08835}{{\normalfont [arXiv:cs,
479 | stat/1902.08835]}}.
480 |
481 | \bibitem[Murray \em{et~al.}(2019)Murray, Stankovic, Stankovic, Lulic, and
482 | Sladojevic]{murray2019}
483 | Murray, D.; Stankovic, L.; Stankovic, V.; Lulic, S.; Sladojevic, S.
484 | \newblock Transferability of {{Neural Network Approaches}} for {{Low}}-Rate
485 | {{Energy Disaggregation}}.
486 | \newblock {{ICASSP}} 2019 - 2019 {{IEEE International Conference}} on
487 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}), 2019, pp.
488 | 8330--8334.
489 | \newblock
490 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP.2019.8682486}{\detokenize{10.1109/ICASSP.2019.8682486}}}.
491 |
492 | \bibitem[Kyrkou \em{et~al.}(2019)Kyrkou, Nalmpantis, and Vrakas]{kyrkou2019}
493 | Kyrkou, L.; Nalmpantis, C.; Vrakas, D.
494 | \newblock Imaging Time-Series for {{Nilm}}.
495 | \newblock International {{Conference}} on {{Engineering Applications}} of
496 | {{Neural Networks}}; {Springer}: {Xersonisos, Greece}, 2019; pp. 188--196.
497 | \newblock
498 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/978-3-030-20257-6_16}{\detokenize{10.1007/978-3-030-20257-6_16}}}.
499 |
500 | \bibitem[Kaselimi \em{et~al.}(2020)Kaselimi, Doulamis, Voulodimos,
501 | Protopapadakis, and Doulamis]{kaselimi2020a}
502 | Kaselimi, M.; Doulamis, N.; Voulodimos, A.; Protopapadakis, E.; Doulamis, A.
503 | \newblock Context {{Aware Energy Disaggregation Using Adaptive Bidirectional
504 | LSTM Models}}.
505 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2020}, {\em
506 | 11},~3054--3067.
507 | \newblock
508 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2020.2974347}{\detokenize{10.1109/TSG.2020.2974347}}}.
509 |
510 | \bibitem[Ahmed \em{et~al.}(2020)Ahmed, Zhang, and Eliassen]{ahmed2020a}
511 | Ahmed, A.M.A.; Zhang, Y.; Eliassen, F.
512 | \newblock Generative {{Adversarial Networks}} and {{Transfer Learning}} for
513 | {{Non}}-{{Intrusive Load Monitoring}} in {{Smart Grids}}.
514 | \newblock 2020 {{IEEE International Conference}} on {{Communications}},
515 | {{Control}}, and {{Computing Technologies}} for {{Smart Grids}}
516 | ({{SmartGridComm}}); {IEEE}: {Tempe, AZ, USA}, 2020; pp. 1--7.
517 | \newblock
518 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SmartGridComm47815.2020.9302933}{\detokenize{10.1109/SmartGridComm47815.2020.9302933}}}.
519 |
520 | \bibitem[Jia \em{et~al.}(2019)Jia, Batra, Wang, and Whitehouse]{jia2019}
521 | Jia, Y.; Batra, N.; Wang, H.; Whitehouse, K.
522 | \newblock A {{Tree}}-{{Structured Neural Network Model}} for {{Household Energy
523 | Breakdown}}.
524 | \newblock The {{World Wide Web Conference}}; {Association for Computing
525 | Machinery}: {San Francisco, CA, USA}, 2019; pp. 2872--2878.
526 | \newblock
527 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3308558.3313405}{\detokenize{10.1145/3308558.3313405}}}.
528 |
529 | \bibitem[Laptev \em{et~al.}(2018)Laptev, Ji, and Rajagopal]{laptev2018}
530 | Laptev, N.; Ji, Y.; Rajagopal, R.
531 | \newblock Using the {{Wisdom}} of {{Neighbors}} for {{Energy Disaggregation}}
532 | from {{Smart Meters}}.
533 | \newblock Proceedings of the 4th {{International Workshop}} on
534 | {{Non}}-{{Intrusive Load Monitoring}}; , 2018; pp. 1--5.
535 |
536 | \bibitem[Shin \em{et~al.}(2019)Shin, Rho, Lee, and Rhee]{shin2019a}
537 | Shin, C.; Rho, S.; Lee, H.; Rhee, W.
538 | \newblock Data {{Requirements}} for {{Applying Machine Learning}} to {{Energy
539 | Disaggregation}}.
540 | \newblock {\em Energies} {\bf 2019}, {\em 12},~1696.
541 | \newblock
542 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en12091696}{\detokenize{10.3390/en12091696}}}.
543 |
544 | \bibitem[Barsim and Yang(2018)]{barsim2018}
545 | Barsim, K.S.; Yang, B.
546 | \newblock On the {{Feasibility}} of {{Generic Deep Disaggregation}} for
547 | {{Single}}-{{Load Extraction}}.
548 | \newblock {{arXiv}} Preprint {{arXiv}}:1802.02139; {arXiv:1802.02139 [cs]}:
549 | {Austin, Texas}, 2018;
550 | \href{http://xxx.lanl.gov/abs/1802.02139}{{\normalfont [1802.02139]}}.
551 |
552 | \bibitem[Rafiq \em{et~al.}(2020)Rafiq, Shi, Zhang, Li, and Ochani]{rafiq2020}
553 | Rafiq, H.; Shi, X.; Zhang, H.; Li, H.; Ochani, M.K.
554 | \newblock A {{Deep Recurrent Neural Network}} for {{Non}}-{{Intrusive Load
555 | Monitoring Based}} on {{Multi}}-{{Feature Input Space}} and
556 | {{Post}}-{{Processing}}.
557 | \newblock {\em Energies} {\bf 2020}, {\em 13},~2195.
558 | \newblock
559 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en13092195}{\detokenize{10.3390/en13092195}}}.
560 |
561 | \bibitem[Kong \em{et~al.}(2020)Kong, Dong, Wang, Zhao, and Huang]{kong2020}
562 | Kong, W.; Dong, Z.Y.; Wang, B.; Zhao, J.; Huang, J.
563 | \newblock A {{Practical Solution}} for {{Non}}-{{Intrusive Type II Load
564 | Monitoring Based}} on {{Deep Learning}} and {{Post}}-{{Processing}}.
565 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2020}, {\em 11},~148--160.
566 | \newblock
567 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2019.2918330}{\detokenize{10.1109/TSG.2019.2918330}}}.
568 |
569 | \bibitem[Huchtkoetter and Reinhardt(2020)]{huchtkoetter2020}
570 | Huchtkoetter, J.; Reinhardt, A.
571 | \newblock On the {{Impact}} of {{Temporal Data Resolution}} on the {{Accuracy}}
572 | of {{Non}}-{{Intrusive Load Monitoring}}.
573 | \newblock Proceedings of the 7th {{ACM International Conference}} on
574 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
575 | {{Transportation}}; {Association for Computing Machinery}: {New York, NY,
576 | USA}, 2020; {{BuildSys}} '20, pp. 270--273.
577 | \newblock
578 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3408308.3427974}{\detokenize{10.1145/3408308.3427974}}}.
579 |
580 | \bibitem[Davies \em{et~al.}(2019)Davies, Dennis, Hansom, Martin, Stankevicius,
581 | and Ward]{davies2019}
582 | Davies, P.; Dennis, J.; Hansom, J.; Martin, W.; Stankevicius, A.; Ward, L.
583 | \newblock Deep {{Neural Networks}} for {{Appliance Transient Classification}}.
584 | \newblock {{ICASSP}} 2019 - 2019 {{IEEE International Conference}} on
585 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}), 2019, pp.
586 | 8320--8324.
587 | \newblock
588 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP.2019.8682658}{\detokenize{10.1109/ICASSP.2019.8682658}}}.
589 |
590 | \bibitem[Murray \em{et~al.}(2020)Murray, Stankovic, and Stankovic]{murray2020}
591 | Murray, D.; Stankovic, L.; Stankovic, V.
592 | \newblock Explainable {{NILM Networks}}.
593 | \newblock Proceedings of the 5th {{International Workshop}} on
594 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
595 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 64--69.
596 | \newblock
597 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427855}{\detokenize{10.1145/3427771.3427855}}}.
598 |
599 | \bibitem[Ayub and {El-Alfy}(2020)]{ayub2020}
600 | Ayub, M.; {El-Alfy}, E.S.M.
601 | \newblock Impact of {{Normalization}} on {{BiLSTM Based Models}} for {{Energy
602 | Disaggregation}}.
603 | \newblock 2020 {{International Conference}} on {{Data Analytics}} for
604 | {{Business}} and {{Industry}}: {{Way Towards}} a {{Sustainable Economy}}
605 | ({{ICDABI}}), 2020, pp. 1--6.
606 | \newblock
607 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICDABI51230.2020.9325593}{\detokenize{10.1109/ICDABI51230.2020.9325593}}}.
608 |
609 | \bibitem[Delfosse \em{et~al.}(2020)Delfosse, {lien}, Hebrail, and
610 | Zerroug]{delfosse2020}
611 | Delfosse, A.; {lien}.; Hebrail, G.; Zerroug, A.
612 | \newblock Deep {{Learning Applied}} to {{NILM}}: {{Is Data Augmentation Worth}}
613 | for {{Energy Disaggregation}}?
614 | \newblock {\em ECAI 2020} {\bf 2020}, pp. 2972--2977.
615 | \newblock
616 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3233/FAIA200471}{\detokenize{10.3233/FAIA200471}}}.
617 |
618 | \bibitem[Pan \em{et~al.}(2020)Pan, Liu, Shen, Cai, and Jia]{pan2020}
619 | Pan, Y.; Liu, K.; Shen, Z.; Cai, X.; Jia, Z.
620 | \newblock Sequence-{{To}}-{{Subsequence Learning With Conditional Gan For Power
621 | Disaggregation}}.
622 | \newblock {{ICASSP}} 2020 - 2020 {{IEEE International Conference}} on
623 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}), 2020, pp.
624 | 3202--3206.
625 | \newblock
626 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP40776.2020.9053947}{\detokenize{10.1109/ICASSP40776.2020.9053947}}}.
627 |
628 | \bibitem[Ulyanov \em{et~al.}(2017)Ulyanov, Vedaldi, and Lempitsky]{ulyanov2017}
629 | Ulyanov, D.; Vedaldi, A.; Lempitsky, V.
630 | \newblock Instance {{Normalization}}: {{The Missing Ingredient}} for {{Fast
631 | Stylization}}.
632 | \newblock {\em arXiv:1607.08022 [cs]} {\bf 2017},
633 | \href{http://xxx.lanl.gov/abs/1607.08022}{{\normalfont
634 | [arXiv:cs/1607.08022]}}.
635 |
636 | \bibitem[Ioffe and Szegedy(2015)]{ioffe2015}
637 | Ioffe, S.; Szegedy, C.
638 | \newblock Batch {{Normalization}}: {{Accelerating Deep Network Training}} by
639 | {{Reducing Internal Covariate Shift}}.
640 | \newblock {\em arXiv:1502.03167 [cs]} {\bf 2015},
641 | \href{http://xxx.lanl.gov/abs/1502.03167}{{\normalfont
642 | [arXiv:cs/1502.03167]}}.
643 |
644 | \bibitem[Bao \em{et~al.}(2018)Bao, Ibrahimov, Wagner, and Schmeck]{bao2018}
645 | Bao, K.; Ibrahimov, K.; Wagner, M.; Schmeck, H.
646 | \newblock Enhancing Neural Non-Intrusive Load Monitoring with Generative
647 | Adversarial Networks.
648 | \newblock {\em Energy Informatics} {\bf 2018}, {\em 1},~18.
649 | \newblock
650 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1186/s42162-018-0038-y}{\detokenize{10.1186/s42162-018-0038-y}}}.
651 |
652 | \bibitem[Chen \em{et~al.}(2018)Chen, Wang, He, Chen, Hu, and He]{chen2018b}
653 | Chen, K.; Wang, Q.; He, Z.; Chen, K.; Hu, J.; He, J.
654 | \newblock Convolutional {{Sequence}} to {{Sequence Non}}-Intrusive {{Load
655 | Monitoring}} ({{arXiv Version}}).
656 | \newblock {\em arXiv:1806.02078 [cs, stat]} {\bf 2018},
657 | \href{http://xxx.lanl.gov/abs/1806.02078}{{\normalfont [arXiv:cs,
658 | stat/1806.02078]}}.
659 |
660 | \bibitem[Chen \em{et~al.}(2020)Chen, Zhang, Wang, Hu, Fan, and He]{chen2020e}
661 | Chen, K.; Zhang, Y.; Wang, Q.; Hu, J.; Fan, H.; He, J.
662 | \newblock Scale- and {{Context}}-{{Aware Convolutional Non}}-{{Intrusive Load
663 | Monitoring}}.
664 | \newblock {\em IEEE Transactions on Power Systems} {\bf 2020}, {\em
665 | 35},~2362--2373.
666 | \newblock
667 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TPWRS.2019.2953225}{\detokenize{10.1109/TPWRS.2019.2953225}}}.
668 |
669 | \bibitem[Kaselimi \em{et~al.}(2019)Kaselimi, Protopapadakis, Voulodimos,
670 | Doulamis, and Doulamis]{kaselimi2019}
671 | Kaselimi, M.; Protopapadakis, E.; Voulodimos, A.; Doulamis, N.; Doulamis, A.
672 | \newblock Multi-{{Channel Recurrent Convolutional Neural Networks}} for
673 | {{Energy Disaggregation}}.
674 | \newblock {\em IEEE Access} {\bf 2019}, {\em 7},~81047--81056.
675 | \newblock
676 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ACCESS.2019.2923742}{\detokenize{10.1109/ACCESS.2019.2923742}}}.
677 |
678 | \bibitem[Liu(2020)]{liu2020}
679 | Liu, H.
680 | \newblock {\em Non-Intrusive {{Load Monitoring}}: {{Theory}}, {{Technologies}}
681 | and {{Applications}}}; {Springer Singapore}: {Singapore}, 2020.
682 | \newblock
683 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/978-981-15-1860-7}{\detokenize{10.1007/978-981-15-1860-7}}}.
684 |
685 | \bibitem[Mottahedi and Asadi(2016)]{mottahedi2016}
686 | Mottahedi, M.; Asadi, S.
687 | \newblock Non-Intrusive {{Load Monitoring Using Imaging Time Series}} and
688 | {{Convolutional Neural Networks}}.
689 | \newblock Proceedings of the {{ICCCBE}} 2016; , 2016.
690 |
691 | \bibitem[Zhang \em{et~al.}(2016)Zhang, Zhong, Wang, Goddard, and
692 | Sutton]{zhang2016}
693 | Zhang, C.; Zhong, M.; Wang, Z.; Goddard, N.; Sutton, C.
694 | \newblock Sequence-to-Point Learning with Neural Networks for Nonintrusive Load
695 | Monitoring.
696 | \newblock {\em arXiv preprint arXiv:1612.09106} {\bf 2016},
697 | \href{http://xxx.lanl.gov/abs/1612.09106}{{\normalfont [1612.09106]}}.
698 |
699 | \bibitem[Shorten and Khoshgoftaar(2019)]{shorten2019}
700 | Shorten, C.; Khoshgoftaar, T.M.
701 | \newblock A Survey on {{Image Data Augmentation}} for {{Deep Learning}}.
702 | \newblock {\em Journal of Big Data} {\bf 2019}, {\em 6},~60.
703 | \newblock
704 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1186/s40537-019-0197-0}{\detokenize{10.1186/s40537-019-0197-0}}}.
705 |
706 | \bibitem[Iwana and Uchida(2020)]{iwana2020}
707 | Iwana, B.K.; Uchida, S.
708 | \newblock An {{Empirical Survey}} of {{Data Augmentation}} for {{Time Series
709 | Classification}} with {{Neural Networks}}.
710 | \newblock {\em arXiv:2007.15951 [cs, stat]} {\bf 2020},
711 | \href{http://xxx.lanl.gov/abs/2007.15951}{{\normalfont [arXiv:cs,
712 | stat/2007.15951]}}.
713 |
714 | \bibitem[Wen \em{et~al.}(2020)Wen, Sun, Song, Gao, Wang, and Xu]{wen2020}
715 | Wen, Q.; Sun, L.; Song, X.; Gao, J.; Wang, X.; Xu, H.
716 | \newblock Time {{Series Data Augmentation}} for {{Deep Learning}}: {{A
717 | Survey}}.
718 | \newblock {\em arXiv:2002.12478 [cs, eess, stat]} {\bf 2020},
719 | \href{http://xxx.lanl.gov/abs/2002.12478}{{\normalfont [arXiv:cs, eess,
720 | stat/2002.12478]}}.
721 |
722 | \bibitem[Garcia \em{et~al.}(2017)Garcia, Creayla, and Macabebe]{garcia2017}
723 | Garcia, F.C.C.; Creayla, C.M.C.; Macabebe, E.Q.B.
724 | \newblock Development of an {{Intelligent System}} for {{Smart Home Energy
725 | Disaggregation Using Stacked Denoising Autoencoders}}.
726 | \newblock {\em Procedia Computer Science} {\bf 2017}, {\em 105},~248--255.
727 | \newblock
728 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.procs.2017.01.218}{\detokenize{10.1016/j.procs.2017.01.218}}}.
729 |
730 | \bibitem[Chang and Ho(2019)]{chang2019}
731 | Chang, F.Y.; Ho, W.J.
732 | \newblock An {{Analysis}} of {{Semi}}-{{Supervised Learning Approaches}} in
733 | {{Low}}-{{Rate Energy Disaggregation}}.
734 | \newblock 2019 3rd {{International Conference}} on {{Smart Grid}} and {{Smart
735 | Cities}} ({{ICSGSC}}), 2019, pp. 145--150.
736 | \newblock
737 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICSGSC.2019.000-4}{\detokenize{10.1109/ICSGSC.2019.000-4}}}.
738 |
739 | \bibitem[Cui \em{et~al.}(2019)Cui, Liu, Luan, and Yu]{cui2019}
740 | Cui, G.; Liu, B.; Luan, W.; Yu, Y.
741 | \newblock Estimation of {{Target Appliance Electricity Consumption Using
742 | Background Filtering}}.
743 | \newblock {\em IEEE Transactions on Smart Grid} {\bf 2019}, {\em
744 | 10},~5920--5929.
745 | \newblock
746 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TSG.2019.2892841}{\detokenize{10.1109/TSG.2019.2892841}}}.
747 |
748 | \bibitem[Reynaud \em{et~al.}(2017)Reynaud, Haradji, Semp{\'e}, and
749 | Sabouret]{reynaud2017}
750 | Reynaud, Q.; Haradji, Y.; Semp{\'e}, F.; Sabouret, N.
751 | \newblock Using {{Time Use Surveys}} in {{Multi Agent}} Based {{Simulations}}
752 | of {{Human Activity}}.
753 | \newblock Proceedings of the 9th {{International Conference}} on {{Agents}}
754 | and {{Artificial Intelligence}}; {SCITEPRESS - Science and Technology
755 | Publications}: {Porto, Portugal}, 2017; Vol.~1, pp. 67--77.
756 | \newblock
757 | doi:{\changeurlcolor{black}\href{https://doi.org/10.5220/0006189100670077}{\detokenize{10.5220/0006189100670077}}}.
758 |
759 | \bibitem[Harell \em{et~al.}(2019)Harell, Makonin, and Baji{\'c}]{harell2019}
760 | Harell, A.; Makonin, S.; Baji{\'c}, I.V.
761 | \newblock Wavenilm: {{A}} Causal Neural Network for Power Disaggregation from
762 | the Complex Power Signal.
763 | \newblock {\em arXiv:1902.08736 [eess]} {\bf 2019},
764 | \href{http://xxx.lanl.gov/abs/1902.08736}{{\normalfont
765 | [arXiv:eess/1902.08736]}}.
766 |
767 | \bibitem[Kundu \em{et~al.}(2018)Kundu, Juvekar, and Davis]{kundu2018}
768 | Kundu, A.; Juvekar, G.P.; Davis, K.
769 | \newblock Deep {{Neural Network Based Non}}-{{Intrusive Load Status
770 | Recognition}}.
771 | \newblock 2018 {{Clemson University Power Systems Conference}} ({{PSC}}),
772 | 2018, pp. 1--6.
773 | \newblock
774 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/PSC.2018.8664063}{\detokenize{10.1109/PSC.2018.8664063}}}.
775 |
776 | \bibitem[Hosseini \em{et~al.}(2019)Hosseini, Henao, Kelouwani, Agbossou, and
777 | Cardenas]{hosseini2019a}
778 | Hosseini, S.; Henao, N.; Kelouwani, S.; Agbossou, K.; Cardenas, A.
779 | \newblock A {{Study}} on {{Markovian}} and {{Deep Learning Based
780 | Architectures}} for {{Household Appliance}}-Level {{Load Modeling}} and
781 | {{Recognition}}.
782 | \newblock 2019 {{IEEE}} 28th {{International Symposium}} on {{Industrial
783 | Electronics}} ({{ISIE}}); {IEEE}: {Vancouver, BC, Canada}, 2019; pp. 35--40.
784 | \newblock
785 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ISIE.2019.8781186}{\detokenize{10.1109/ISIE.2019.8781186}}}.
786 |
787 | \bibitem[Shin \em{et~al.}(2018)Shin, Joo, Yim, Lee, Moon, and Rhee]{shin2018}
788 | Shin, C.; Joo, S.; Yim, J.; Lee, H.; Moon, T.; Rhee, W.
789 | \newblock Subtask {{Gated Networks}} for {{Non}}-{{Intrusive Load Monitoring}}.
790 | \newblock {\em arXiv:1811.06692 [cs, stat]} {\bf 2018},
791 | \href{http://xxx.lanl.gov/abs/1811.06692}{{\normalfont [arXiv:cs,
792 | stat/1811.06692]}}.
793 |
794 | \bibitem[Jiang \em{et~al.}(2019)Jiang, Kong, Plumbley, and Gilbert]{jiang2019}
795 | Jiang, J.; Kong, Q.; Plumbley, M.; Gilbert, N.
796 | \newblock Deep {{Learning Based Energy Disaggregation}} and {{On}}/{{Off
797 | Detection}} of {{Household Appliances}}.
798 | \newblock {\em arXiv:1908.00941 [cs, eess]} {\bf 2019},
799 | \href{http://xxx.lanl.gov/abs/1908.00941}{{\normalfont [arXiv:cs,
800 | eess/1908.00941]}}.
801 |
802 | \bibitem[Reinhardt and Bouchur(2020)]{reinhardt2020a}
803 | Reinhardt, A.; Bouchur, M.
804 | \newblock On the {{Impact}} of the {{Sequence Length}} on
805 | {{Sequence}}-to-{{Sequence}} and {{Sequence}}-to-{{Point Learning}} for
806 | {{NILM}}.
807 | \newblock Proceedings of the 5th {{International Workshop}} on
808 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
809 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 75--78.
810 | \newblock
811 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427857}{\detokenize{10.1145/3427771.3427857}}}.
812 |
813 | \bibitem[Bousbiat \em{et~al.}(2020)Bousbiat, Klemenjak, and
814 | Elmenreich]{bousbiat2020}
815 | Bousbiat, H.; Klemenjak, C.; Elmenreich, W.
816 | \newblock Exploring {{Time Series Imaging}} for {{Load Disaggregation}}.
817 | \newblock Proceedings of the 7th {{ACM International Conference}} on
818 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
819 | {{Transportation}}; {Association for Computing Machinery}: {New York, NY,
820 | USA}, 2020; {{BuildSys}} '20, pp. 254--257.
821 | \newblock
822 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3408308.3427975}{\detokenize{10.1145/3408308.3427975}}}.
823 |
824 | \bibitem[Wang and Oates(2015)]{wang2015}
825 | Wang, Z.; Oates, T.
826 | \newblock Encoding Time Series as Images for Visual Inspection and
827 | Classification Using Tiled Convolutional Neural Networks.
828 | \newblock Workshops at the Twenty-Ninth {{AAAI}} Conference on Artificial
829 | Intelligence; {AAAI Publications}: {Austin, Texas, USA}, 2015; Vol.~1.
830 |
831 | \bibitem[Eckmann \em{et~al.}(1987)Eckmann, Kamphorst, and Ruelle]{eckmann1995}
832 | Eckmann, J.P.; Kamphorst, S.O.; Ruelle, D.
833 | \newblock Recurrence {{Plots}} of {{Dynamical Systems}}.
834 | \newblock {\em Europhysics Letters (EPL)} {\bf 1987}, {\em 4},~973--977.
835 | \newblock
836 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1209/0295-5075/4/9/004}{\detokenize{10.1209/0295-5075/4/9/004}}}.
837 |
838 | \bibitem[{de Paiva Penha} and Castro(2017)]{depaivapenha2017}
839 | {de Paiva Penha}, D.; Castro, A.R.G.
840 | \newblock Convolutional Neural Network Applied to the Identification of
841 | Residential Equipment in Non-Intrusive Load Monitoring Systems.
842 | \newblock Computer {{Science}} \& {{Information Technology}}, {{CSCP}} 2017.
843 | {AIRCC Digital Library}, 2017, pp. 11--21.
844 | \newblock
845 | doi:{\changeurlcolor{black}\href{https://doi.org/10.5121/csit.2017.71802}{\detokenize{10.5121/csit.2017.71802}}}.
846 |
847 | \bibitem[{de Paiva Penha} and Castro(2018)]{depaivapenha2018}
848 | {de Paiva Penha}, D.; Castro, A.R.G.
849 | \newblock Home Appliance Identification for {{NILM}} Systems Based on Deep
850 | Neural Networks.
851 | \newblock {\em International Journal of Artificial Intelligence and
852 | Applications (IJAIA)} {\bf 2018}, {\em 9},~69--80.
853 | \newblock
854 | doi:{\changeurlcolor{black}\href{https://doi.org/10.5121/ijaia.2018.9206}{\detokenize{10.5121/ijaia.2018.9206}}}.
855 |
856 | \bibitem[Li \em{et~al.}(2019)Li, Zheng, Liu, and Zhang]{li2019a}
857 | Li, C.; Zheng, R.; Liu, M.; Zhang, S.
858 | \newblock A {{Fusion Framework}} Using {{Integrated Neural Network Model}} for
859 | {{Non}}-{{Intrusive Load Monitoring}}.
860 | \newblock 2019 {{Chinese Control Conference}} ({{CCC}}); {IEEE}: {Guangzhou,
861 | China}, 2019; pp. 7385--7390.
862 | \newblock
863 | doi:{\changeurlcolor{black}\href{https://doi.org/10.23919/ChiCC.2019.8865721}{\detokenize{10.23919/ChiCC.2019.8865721}}}.
864 |
865 | \bibitem[Buchhop and Ranganathan(2019)]{buchhop2019}
866 | Buchhop, S.J.; Ranganathan, P.
867 | \newblock Residential {{Load Identification Based}} on {{Load Profile}} Using
868 | {{Artificial Neural Network}} ({{ANN}}).
869 | \newblock 2019 {{North American Power Symposium}} ({{NAPS}}); {IEEE}:
870 | {Wichita, KS, USA}, 2019; pp. 1--6.
871 | \newblock
872 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/NAPS46351.2019.9000373}{\detokenize{10.1109/NAPS46351.2019.9000373}}}.
873 |
874 | \bibitem[LeCun \em{et~al.}(2015)LeCun, Bengio, and Hinton]{lecun2015}
875 | LeCun, Y.; Bengio, Y.; Hinton, G.
876 | \newblock Deep Learning.
877 | \newblock {\em Nature} {\bf 2015}, {\em 521},~436--444.
878 | \newblock
879 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/nature14539}{\detokenize{10.1038/nature14539}}}.
880 |
881 | \bibitem[Vincent \em{et~al.}(2008)Vincent, Larochelle, Bengio, and
882 | Manzagol]{vincent2008}
883 | Vincent, P.; Larochelle, H.; Bengio, Y.; Manzagol, P.A.
884 | \newblock Extracting and Composing Robust Features with Denoising Autoencoders.
885 | \newblock Proceedings of the 25th International Conference on {{Machine}}
886 | Learning; {Association for Computing Machinery}: {New York, NY, USA}, 2008;
887 | {{ICML}} '08, pp. 1096--1103.
888 | \newblock
889 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/1390156.1390294}{\detokenize{10.1145/1390156.1390294}}}.
890 |
891 | \bibitem[Kingma and Welling(2014)]{kingma2014a}
892 | Kingma, D.P.; Welling, M.
893 | \newblock Auto-{{Encoding Variational Bayes}}.
894 | \newblock {\em arXiv:1312.6114 [cs, stat]} {\bf 2014},
895 | \href{http://xxx.lanl.gov/abs/1312.6114}{{\normalfont [arXiv:cs,
896 | stat/1312.6114]}}.
897 |
898 | \bibitem[Rezende \em{et~al.}(2014)Rezende, Mohamed, and Wierstra]{rezende2014}
899 | Rezende, D.J.; Mohamed, S.; Wierstra, D.
900 | \newblock Stochastic {{Backpropagation}} and {{Approximate Inference}} in
901 | {{Deep Generative Models}}.
902 | \newblock {\em arXiv:1401.4082 [cs, stat]} {\bf 2014},
903 | \href{http://xxx.lanl.gov/abs/1401.4082}{{\normalfont [arXiv:cs,
904 | stat/1401.4082]}}.
905 |
906 | \bibitem[van~den Oord \em{et~al.}(2016)van~den Oord, Dieleman, Zen, Simonyan,
907 | Vinyals, Graves, Kalchbrenner, Senior, and Kavukcuoglu]{oord2016}
908 | van~den Oord, A.; Dieleman, S.; Zen, H.; Simonyan, K.; Vinyals, O.; Graves, A.;
909 | Kalchbrenner, N.; Senior, A.; Kavukcuoglu, K.
910 | \newblock {{WaveNet}}: {{A Generative Model}} for {{Raw Audio}}.
911 | \newblock {\em arXiv:1609.03499 [cs]} {\bf 2016},
912 | \href{http://xxx.lanl.gov/abs/1609.03499}{{\normalfont
913 | [arXiv:cs/1609.03499]}}.
914 |
915 | \bibitem[Yue \em{et~al.}(18.11. 2020)Yue, Witzig, Jorde, and Jacobsen]{yue2020}
916 | Yue, Z.; Witzig, C.R.; Jorde, D.; Jacobsen, H.A.
917 | \newblock {{BERT4NILM}}: {{A Bidirectional Transformer Model}} for
918 | {{Non}}-{{Intrusive Load Monitoring}}.
919 | \newblock Proceedings of the 5th {{International Workshop}} on
920 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
921 | {Yokohama, Japan}, 18.11. 2020; {{NILM}}'20, pp. 89--93.
922 | \newblock
923 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3429390}{\detokenize{10.1145/3427771.3429390}}}.
924 |
925 | \bibitem[Devlin \em{et~al.}(2018)Devlin, Chang, Lee, and Toutanova]{devlin2018}
926 | Devlin, J.; Chang, M.W.; Lee, K.; Toutanova, K.
927 | \newblock Bert: {{Pre}}-Training of Deep Bidirectional Transformers for
928 | Language Understanding.
929 | \newblock {\em arXiv preprint arXiv:1810.04805} {\bf 2018},
930 | \href{http://xxx.lanl.gov/abs/1810.04805}{{\normalfont [1810.04805]}}.
931 |
932 | \bibitem[Zhang \em{et~al.}(2019)Zhang, Goodfellow, Metaxas, and
933 | Odena]{zhang2019a}
934 | Zhang, H.; Goodfellow, I.; Metaxas, D.; Odena, A.
935 | \newblock Self-{{Attention Generative Adversarial Networks}}.
936 | \newblock Proceedings of the 36th {{International Conference}} on {{Machine
937 | Learning}}; {PMLR}: {Long Beach, California, USA}, 2019; Vol.~97, pp.
938 | 7354--7363.
939 |
940 | \bibitem[Sudoso and Piccialli(2019)]{sudoso2019}
941 | Sudoso, A.M.; Piccialli, V.
942 | \newblock Non-{{Intrusive Load Monitoring}} with an {{Attention}}-Based {{Deep
943 | Neural Network}}.
944 | \newblock {\em arXiv:1912.00759 [cs, eess, stat]} {\bf 2019},
945 | \href{http://xxx.lanl.gov/abs/1912.00759}{{\normalfont [arXiv:cs, eess,
946 | stat/1912.00759]}}.
947 |
948 | \bibitem[Raffel and Ellis(2016)]{raffel2016}
949 | Raffel, C.; Ellis, D.P.W.
950 | \newblock Feed-{{Forward Networks}} with {{Attention Can Solve Some
951 | Long}}-{{Term Memory Problems}}.
952 | \newblock {\em arXiv:1512.08756 [cs]} {\bf 2016},
953 | \href{http://xxx.lanl.gov/abs/1512.08756}{{\normalfont
954 | [arXiv:cs/1512.08756]}}.
955 |
956 | \bibitem[Hochreiter and Schmidhuber(1997)]{hochreiter1997}
957 | Hochreiter, S.; Schmidhuber, J.
958 | \newblock Long {{Short}}-{{Term Memory}}.
959 | \newblock {\em Neural Computation} {\bf 1997}, {\em 9},~1735--1780.
960 | \newblock
961 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1162/neco.1997.9.8.1735}{\detokenize{10.1162/neco.1997.9.8.1735}}}.
962 |
963 | \bibitem[Graves and Schmidhuber(2005)]{graves2005}
964 | Graves, A.; Schmidhuber, J.
965 | \newblock Framewise Phoneme Classification with Bidirectional {{LSTM}} and
966 | Other Neural Network Architectures.
967 | \newblock {\em Neural Networks} {\bf 2005}, {\em 18},~602--610.
968 | \newblock
969 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.neunet.2005.06.042}{\detokenize{10.1016/j.neunet.2005.06.042}}}.
970 |
971 | \bibitem[Cho \em{et~al.}(2014)Cho, {van Merrienboer}, Gulcehre, Bahdanau,
972 | Bougares, Schwenk, and Bengio]{cho2014}
973 | Cho, K.; {van Merrienboer}, B.; Gulcehre, C.; Bahdanau, D.; Bougares, F.;
974 | Schwenk, H.; Bengio, Y.
975 | \newblock Learning {{Phrase Representations}} Using {{RNN Encoder}}-{{Decoder}}
976 | for {{Statistical Machine Translation}}.
977 | \newblock {\em arXiv:1406.1078 [cs, stat]} {\bf 2014},
978 | \href{http://xxx.lanl.gov/abs/1406.1078}{{\normalfont [arXiv:cs,
979 | stat/1406.1078]}}.
980 |
981 | \bibitem[Goodfellow \em{et~al.}(2014)Goodfellow, {Pouget-Abadie}, Mirza, Xu,
982 | {Warde-Farley}, Ozair, Courville, and Bengio]{goodfellow2014}
983 | Goodfellow, I.; {Pouget-Abadie}, J.; Mirza, M.; Xu, B.; {Warde-Farley}, D.;
984 | Ozair, S.; Courville, A.; Bengio, Y.
985 | \newblock Generative Adversarial Nets.
986 | \newblock Advances in Neural Information Processing Systems; Ghahramani, Z.;
987 | Welling, M.; Cortes, C.; Lawrence, N.; Weinberger, K.Q., Eds. {Curran
988 | Associates, Inc.}, 2014, Vol.~27, pp. 2672--2680.
989 |
990 | \bibitem[Liang and Hu(2015)]{liang2015}
991 | Liang, M.; Hu, X.
992 | \newblock Recurrent {{Convolutional Neural Network}} for {{Object
993 | Recognition}}.
994 | \newblock Proceedings of the {{IEEE Conference}} on {{Computer Vision}} and
995 | {{Pattern Recognition}}, 2015, pp. 3367--3375.
996 |
997 | \bibitem[Mauch and Yang(2016)]{mauch2016}
998 | Mauch, L.; Yang, B.
999 | \newblock A Novel {{DNN}}-{{HMM}}-Based Approach for Extracting Single Loads
1000 | from Aggregate Power Signals.
1001 | \newblock 2016 {{IEEE International Conference}} on {{Acoustics}}, {{Speech}}
1002 | and {{Signal Processing}} ({{ICASSP}}), 2016, pp. 2384--2388.
1003 | \newblock
1004 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP.2016.7472104}{\detokenize{10.1109/ICASSP.2016.7472104}}}.
1005 |
1006 | \bibitem[Chung \em{et~al.}(2015)Chung, Kastner, Dinh, Goel, Courville, and
1007 | Bengio]{chung2015}
1008 | Chung, J.; Kastner, K.; Dinh, L.; Goel, K.; Courville, A.C.; Bengio, Y.
1009 | \newblock A Recurrent Latent Variable Model for Sequential Data.
1010 | \newblock Advances in Neural Information Processing Systems. {Curran
1011 | Associates, Inc.}, 2015, Vol.~28.
1012 |
1013 | \bibitem[Chen \em{et~al.}(2020)Chen, Wang, and Fan]{chen2020c}
1014 | Chen, H.; Wang, Y.H.; Fan, C.H.
1015 | \newblock A Convolutional Autoencoder-Based Approach with Batch Normalization
1016 | for Energy Disaggregation.
1017 | \newblock {\em The Journal of Supercomputing} {\bf 2020}, {\em 77},~2961--2978.
1018 | \newblock
1019 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/s11227-020-03375-y}{\detokenize{10.1007/s11227-020-03375-y}}}.
1020 |
1021 | \bibitem[Kaselimi \em{et~al.}(2019)Kaselimi, Doulamis, Doulamis, Voulodimos,
1022 | and Protopapadakis]{kaselimi2019a}
1023 | Kaselimi, M.; Doulamis, N.; Doulamis, A.; Voulodimos, A.; Protopapadakis, E.
1024 | \newblock Bayesian-Optimized {{Bidirectional LSTM Regression Model}} for
1025 | {{Non}}-Intrusive {{Load Monitoring}}.
1026 | \newblock {{ICASSP}} 2019 - 2019 {{IEEE International Conference}} on
1027 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}); {IEEE}:
1028 | {Brighton, United Kingdom}, 2019; pp. 2747--2751.
1029 | \newblock
1030 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP.2019.8683110}{\detokenize{10.1109/ICASSP.2019.8683110}}}.
1031 |
1032 | \bibitem[Dash and Naik(2018)]{dash2018}
1033 | Dash, P.; Naik, K.
1034 | \newblock A {{Very Deep One Dimensional Convolutional Neural Network}}
1035 | ({{VDOCNN}}) for {{Appliance Power Signature Classification}}.
1036 | \newblock 2018 {{IEEE Electrical Power}} and {{Energy Conference}} ({{EPEC}}),
1037 | 2018, pp. 1--6.
1038 | \newblock
1039 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/EPEC.2018.8598355}{\detokenize{10.1109/EPEC.2018.8598355}}}.
1040 |
1041 | \bibitem[Nascimento(2016)]{nascimento2016}
1042 | Nascimento, P.P.M.
1043 | \newblock Applications of Deep Learning Techniques on Nilm.
1044 | \newblock Master's {{Thesis}}, Universidade Federal do Rio de Janeiro, {Rio de
1045 | Janeiro}, 2016.
1046 |
1047 | \bibitem[Bengio \em{et~al.}(2009)Bengio, Louradour, Collobert, and
1048 | Weston]{bengio2009a}
1049 | Bengio, Y.; Louradour, J.; Collobert, R.; Weston, J.
1050 | \newblock Curriculum Learning.
1051 | \newblock Proceedings of the 26th {{Annual International Conference}} on
1052 | {{Machine Learning}}; {Association for Computing Machinery}: {New York, NY,
1053 | USA}, 2009; {{ICML}} '09, pp. 41--48.
1054 | \newblock
1055 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/1553374.1553380}{\detokenize{10.1145/1553374.1553380}}}.
1056 |
1057 | \bibitem[Koenker and Bassett(1978)]{koenker1978}
1058 | Koenker, R.; Bassett, G.
1059 | \newblock Regression {{Quantiles}}.
1060 | \newblock {\em Econometrica} {\bf 1978}, {\em 46},~33--50.
1061 | \newblock
1062 | doi:{\changeurlcolor{black}\href{https://doi.org/10.2307/1913643}{\detokenize{10.2307/1913643}}}.
1063 |
1064 | \bibitem[Gomes and Pereira(2020)]{gomes2020}
1065 | Gomes, E.; Pereira, L.
1066 | \newblock {{PB}}-{{NILM}}: {{Pinball Guided Deep Non}}-{{Intrusive Load
1067 | Monitoring}}.
1068 | \newblock {\em IEEE Access} {\bf 2020}, {\em 8},~48386--48398.
1069 | \newblock
1070 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ACCESS.2020.2978513}{\detokenize{10.1109/ACCESS.2020.2978513}}}.
1071 |
1072 | \bibitem[Faustine \em{et~al.}(2020)Faustine, Pereira, Bousbiat, and
1073 | Kulkarni]{faustine2020a}
1074 | Faustine, A.; Pereira, L.; Bousbiat, H.; Kulkarni, S.
1075 | \newblock {{UNet}}-{{NILM}}: {{A Deep Neural Network}} for {{Multi}}-Tasks
1076 | {{Appliances State Detection}} and {{Power Estimation}} in {{NILM}}.
1077 | \newblock Proceedings of the 5th {{International Workshop}} on
1078 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
1079 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 84--88.
1080 | \newblock
1081 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427859}{\detokenize{10.1145/3427771.3427859}}}.
1082 |
1083 | \bibitem[Kaselimi \em{et~al.}(2020)Kaselimi, Voulodimos, Protopapadakis,
1084 | Doulamis, and Doulamis]{kaselimi2020}
1085 | Kaselimi, M.; Voulodimos, A.; Protopapadakis, E.; Doulamis, N.; Doulamis, A.
1086 | \newblock {{EnerGAN}}: {{A Generative Adversarial Network}} for {{Energy
1087 | Disaggregation}}.
1088 | \newblock {{ICASSP}} 2020 - 2020 {{IEEE International Conference}} on
1089 | {{Acoustics}}, {{Speech}} and {{Signal Processing}} ({{ICASSP}}), 2020, pp.
1090 | 1578--1582.
1091 | \newblock
1092 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASSP40776.2020.9054342}{\detokenize{10.1109/ICASSP40776.2020.9054342}}}.
1093 |
1094 | \bibitem[Hsu \em{et~al.}(2019)Hsu, Zeitoun, Lee, Katabi, and Jaakkola]{hsu2019}
1095 | Hsu, C.Y.; Zeitoun, A.; Lee, G.H.; Katabi, D.; Jaakkola, T.
1096 | \newblock Self-{{Supervised Learning}} of {{Appliance Usage}}.
1097 | \newblock International {{Conference}} on {{Learning Representations}} -
1098 | {{ICLR}} 2020; , 2019.
1099 |
1100 | \bibitem[Xiao and Cheng(2019)]{xiao2019}
1101 | Xiao, P.; Cheng, S.
1102 | \newblock Neural {{Network}} for {{NILM Based}} on {{Operational State Change
1103 | Classification}}.
1104 | \newblock {\em arXiv:1902.02675 [cs, stat]} {\bf 2019},
1105 | \href{http://xxx.lanl.gov/abs/1902.02675}{{\normalfont [arXiv:cs,
1106 | stat/1902.02675]}}.
1107 |
1108 | \bibitem[Al~Zeidi(2018)]{alzeidi2018}
1109 | Al~Zeidi, O.
1110 | \newblock Deep {{Neural Networks}} for {{Non}}-{{Intrusive Load Monitoring}}.
1111 | \newblock Master's {{Thesis}}, Monash University, 2018.
1112 |
1113 | \bibitem[Wang \em{et~al.}(2019)Wang, Kababji, Graham, and Srikantha]{wang2019b}
1114 | Wang, J.; Kababji, S.E.; Graham, C.; Srikantha, P.
1115 | \newblock Ensemble-{{Based Deep Learning Model}} for {{Non}}-{{Intrusive Load
1116 | Monitoring}}.
1117 | \newblock 2019 {{IEEE Electrical Power}} and {{Energy Conference}} ({{EPEC}}),
1118 | 2019, pp. 1--6.
1119 | \newblock
1120 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/EPEC47565.2019.9074816}{\detokenize{10.1109/EPEC47565.2019.9074816}}}.
1121 |
1122 | \bibitem[Quek \em{et~al.}(2020)Quek, Woo, and Logenthiran]{quek2020}
1123 | Quek, Y.T.; Woo, W.L.; Logenthiran, T.
1124 | \newblock Load {{Disaggregation Using One}}-{{Directional Convolutional Stacked
1125 | Long Short}}-{{Term Memory Recurrent Neural Network}}.
1126 | \newblock {\em IEEE Systems Journal} {\bf 2020}, {\em 14},~1395--1404.
1127 | \newblock
1128 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/JSYST.2019.2919668}{\detokenize{10.1109/JSYST.2019.2919668}}}.
1129 |
1130 | \bibitem[Harell \em{et~al.}(2018)Harell, Makonin, and Bajic]{harell2018}
1131 | Harell, A.; Makonin, S.; Bajic, I.V.
1132 | \newblock A Recurrent Neural Network for Multisensory Non-Intrusive Load
1133 | Monitoring on a {{Raspberry Pi}}.
1134 | \newblock {{IEEE}} 20th {{International Workshop}} on {{Multimedia Signal
1135 | Processing}} ({{MMSP}} 2018); , 2018.
1136 |
1137 | \bibitem[Morgan(2017)]{morgan2017}
1138 | Morgan, E.S.
1139 | \newblock Applications of Deep Learning for Load Disaggregation in Residential
1140 | Environments.
1141 | \newblock Bachelor's thesis, Universidade Federal do Rio de Janeiro, {Rio de
1142 | Janeiro}, 2017.
1143 |
1144 | \bibitem[Massidda \em{et~al.}(2020)Massidda, Marrocu, and Manca]{massidda2020}
1145 | Massidda, L.; Marrocu, M.; Manca, S.
1146 | \newblock Non-{{Intrusive Load Disaggregation}} by {{Convolutional Neural
1147 | Network}} and {{Multilabel Classification}}.
1148 | \newblock {\em Applied Sciences} {\bf 2020}, {\em 10},~1454.
1149 | \newblock
1150 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/app10041454}{\detokenize{10.3390/app10041454}}}.
1151 |
1152 | \bibitem[Kim \em{et~al.}(2017)Kim, Le, and Kim]{kim2017}
1153 | Kim, J.; Le, T.T.H.; Kim, H.
1154 | \newblock Nonintrusive {{Load Monitoring Based}} on {{Advanced Deep Learning}}
1155 | and {{Novel Signature}}.
1156 | \newblock {\em Computational Intelligence and Neuroscience} {\bf 2017}, {\em
1157 | 2017},~4216281.
1158 | \newblock
1159 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1155/2017/4216281}{\detokenize{10.1155/2017/4216281}}}.
1160 |
1161 | \bibitem[Yadav \em{et~al.}(2020)Yadav, Sinha, Saidi, Trinkl, and
1162 | Z{\"o}rner]{yadav2020}
1163 | Yadav, A.; Sinha, A.; Saidi, A.; Trinkl, C.; Z{\"o}rner, W.
1164 | \newblock {{NILM}} Based {{Energy Disaggregation Algorithm}} for {{Dairy
1165 | Farms}}.
1166 | \newblock Proceedings of the 5th {{International Workshop}} on
1167 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
1168 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 16--19.
1169 | \newblock
1170 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427846}{\detokenize{10.1145/3427771.3427846}}}.
1171 |
1172 | \bibitem[Ahmed and Bons(2020)]{ahmed2020}
1173 | Ahmed, S.; Bons, M.
1174 | \newblock Edge Computed {{NILM}}: A Phone-Based Implementation Using
1175 | {{MobileNet}} Compressed by {{Tensorflow Lite}}.
1176 | \newblock Proceedings of the 5th {{International Workshop}} on
1177 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
1178 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 44--48.
1179 | \newblock
1180 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427852}{\detokenize{10.1145/3427771.3427852}}}.
1181 |
1182 | \bibitem[Barber \em{et~al.}(2020)Barber, Cuay{\'a}huitl, Zhong, and
1183 | Luan]{barber2020}
1184 | Barber, J.; Cuay{\'a}huitl, H.; Zhong, M.; Luan, W.
1185 | \newblock Lightweight {{Non}}-{{Intrusive Load Monitoring Employing Pruned
1186 | Sequence}}-to-{{Point Learning}}.
1187 | \newblock Proceedings of the 5th {{International Workshop}} on
1188 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
1189 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 11--15.
1190 | \newblock
1191 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3427845}{\detokenize{10.1145/3427771.3427845}}}.
1192 |
1193 | \bibitem[Kukunuri \em{et~al.}(2020)Kukunuri, Aglawe, Chauhan, Bhagtani, Patil,
1194 | Walia, and Batra]{kukunuri2020}
1195 | Kukunuri, R.; Aglawe, A.; Chauhan, J.; Bhagtani, K.; Patil, R.; Walia, S.;
1196 | Batra, N.
1197 | \newblock {{EdgeNILM}}: {{Towards NILM}} on {{Edge}} Devices.
1198 | \newblock Proceedings of the 7th {{ACM International Conference}} on
1199 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
1200 | {{Transportation}}; {Association for Computing Machinery}: {New York, NY,
1201 | USA}, 2020; {{BuildSys}} '20, pp. 90--99.
1202 | \newblock
1203 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3408308.3427977}{\detokenize{10.1145/3408308.3427977}}}.
1204 |
1205 | \bibitem[Jones \em{et~al.}(2020)Jones, Klemenjak, Makonin, and
1206 | Baji{\'c}]{jones2020}
1207 | Jones, R.; Klemenjak, C.; Makonin, S.; Baji{\'c}, I.V.
1208 | \newblock Stop: {{Exploring Bayesian Surprise}} to {{Better Train NILM}}.
1209 | \newblock Proceedings of the 5th {{International Workshop}} on
1210 | {{Non}}-{{Intrusive Load Monitoring}}; {Association for Computing Machinery}:
1211 | {Yokohama, Japan}, 2020; {{NILM}}'20, pp. 39--43.
1212 | \newblock
1213 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3427771.3429388}{\detokenize{10.1145/3427771.3429388}}}.
1214 |
1215 | \bibitem[Tongta and Chooruang(2020)]{tongta2020}
1216 | Tongta, A.; Chooruang, K.
1217 | \newblock Long {{Short}}-{{Term Memory}} ({{LSTM}}) {{Neural Networks Applied}}
1218 | to {{Energy Disaggregation}}.
1219 | \newblock 2020 8th {{International Electrical Engineering Congress}}
1220 | ({{iEECON}}), 2020, pp. 1--4.
1221 | \newblock
1222 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/iEECON48109.2020.229559}{\detokenize{10.1109/iEECON48109.2020.229559}}}.
1223 |
1224 | \bibitem[Zhang \em{et~al.}(2020)Zhang, Yin, Cong, and Du]{zhang2020}
1225 | Zhang, Y.; Yin, B.; Cong, Y.; Du, Z.
1226 | \newblock Multi-{{State Household Appliance Identification Based}} on
1227 | {{Convolutional Neural Networks}} and {{Clustering}}.
1228 | \newblock {\em Energies} {\bf 2020}, {\em 13},~792.
1229 | \newblock
1230 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en13040792}{\detokenize{10.3390/en13040792}}}.
1231 |
1232 | \bibitem[Xia \em{et~al.}(2020)Xia, Liu, Wang, Song, Chen, and Li]{xia2020}
1233 | Xia, M.; Liu, W.; Wang, K.; Song, W.; Chen, C.; Li, Y.
1234 | \newblock Non-Intrusive Load Disaggregation Based on Composite Deep Long
1235 | Short-Term Memory Network.
1236 | \newblock {\em Expert Systems with Applications} {\bf 2020}, {\em 160},~113669.
1237 | \newblock
1238 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.eswa.2020.113669}{\detokenize{10.1016/j.eswa.2020.113669}}}.
1239 |
1240 | \bibitem[{\c C}avdar and Faryad(2019)]{cavdar2019}
1241 | {\c C}avdar, {\.I}.; Faryad, V.
1242 | \newblock New {{Design}} of a {{Supervised Energy Disaggregation Model Based}}
1243 | on the {{Deep Neural Network}} for a {{Smart Grid}}.
1244 | \newblock {\em Energies} {\bf 2019}, {\em 12},~1217.
1245 | \newblock
1246 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en12071217}{\detokenize{10.3390/en12071217}}}.
1247 |
1248 | \bibitem[Santos \em{et~al.}(2019)Santos, Freitas, and Aquino]{santos2019}
1249 | Santos, E.G.; Freitas, C.G.S.; Aquino, A.L.L.
1250 | \newblock A {{Deep Learning}} Approach for {{Energy Disaggregation}}
1251 | Considering {{Embedded Devices}}.
1252 | \newblock 2019 {{IX Brazilian Symposium}} on {{Computing Systems Engineering}}
1253 | ({{SBESC}}), 2019, pp. 1--8.
1254 | \newblock
1255 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SBESC49506.2019.9046095}{\detokenize{10.1109/SBESC49506.2019.9046095}}}.
1256 |
1257 | \bibitem[Miao \em{et~al.}(2019)Miao, Zhao, Shi, and Zhang]{miao2019}
1258 | Miao, N.; Zhao, S.; Shi, Q.; Zhang, R.
1259 | \newblock Non-{{Intrusive Load Disaggregation Using Semi}}-{{Supervised
1260 | Learning Method}}.
1261 | \newblock 2019 {{International Conference}} on {{Security}}, {{Pattern
1262 | Analysis}}, and {{Cybernetics}} ({{SPAC}}), 2019, pp. 17--22.
1263 | \newblock
1264 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SPAC49953.2019.237865}{\detokenize{10.1109/SPAC49953.2019.237865}}}.
1265 |
1266 | \bibitem[Min \em{et~al.}(2019)Min, Wen, Yang, Li, and Li]{min2019}
1267 | Min, C.; Wen, G.; Yang, Z.; Li, X.; Li, B.
1268 | \newblock Non-{{Intrusive Load Monitoring System Based}} on {{Convolution
1269 | Neural Network}} and {{Adaptive Linear Programming Boosting}}.
1270 | \newblock {\em Energies} {\bf 2019}, {\em 12},~2882.
1271 | \newblock
1272 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en12152882}{\detokenize{10.3390/en12152882}}}.
1273 |
1274 | \bibitem[Jasi{\'n}ski(2019)]{jasinski2019}
1275 | Jasi{\'n}ski, T.
1276 | \newblock Modelling the Disaggregated Demand for Electricity at the Level of
1277 | Residential Buildings with the Use of Artificial Neural Networks (Deep
1278 | Learning Approach).
1279 | \newblock {\em MATEC Web of Conferences} {\bf 2019}, {\em 282},~02077.
1280 | \newblock
1281 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1051/matecconf/201928202077}{\detokenize{10.1051/matecconf/201928202077}}}.
1282 |
1283 | \bibitem[Cao \em{et~al.}(2019)Cao, Weng, Xia, and Zhang]{cao2019}
1284 | Cao, H.; Weng, L.; Xia, M.; Zhang, D.
1285 | \newblock Non-{{Intrusive Load Disaggregation Based}} on {{Residual Gated
1286 | Network}}.
1287 | \newblock {\em IOP Conference Series: Materials Science and Engineering} {\bf
1288 | 2019}, {\em 677},~032092.
1289 | \newblock
1290 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1088/1757-899X/677/3/032092}{\detokenize{10.1088/1757-899X/677/3/032092}}}.
1291 |
1292 | \bibitem[Linh and Arboleya(2019)]{linh2019}
1293 | Linh, N.V.; Arboleya, P.
1294 | \newblock Deep {{Learning Application}} to {{Non}}-{{Intrusive Load
1295 | Monitoring}}.
1296 | \newblock 2019 {{IEEE Milan PowerTech}}; {IEEE}: {Milan, Italy}, 2019; pp.
1297 | 1--5.
1298 | \newblock
1299 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/PTC.2019.8810435}{\detokenize{10.1109/PTC.2019.8810435}}}.
1300 |
1301 | \bibitem[Gopu \em{et~al.}(2019)Gopu, Gudimallam, B, Thokala, and
1302 | Chandra]{gopu2019a}
1303 | Gopu, R.; Gudimallam, A.; B, V.; Thokala, N.; Chandra, M.G.
1304 | \newblock On {{Electrical Load Disaggregation}} Using {{Recurrent Neural
1305 | Networks}}.
1306 | \newblock Proceedings of the 6th {{ACM International Conference}} on
1307 | {{Systems}} for {{Energy}}-{{Efficient Buildings}}, {{Cities}}, and
1308 | {{Transportation}}; {Association for Computing Machinery}: {New York, NY,
1309 | USA}, 2019; {{BuildSys}} '19, pp. 364--365.
1310 | \newblock
1311 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3360322.3361002}{\detokenize{10.1145/3360322.3361002}}}.
1312 |
1313 | \bibitem[Yang \em{et~al.}(2019)Yang, Zhong, Li, Gulliver, and Li]{yang2019}
1314 | Yang, Y.; Zhong, J.; Li, W.; Gulliver, T.A.; Li, S.
1315 | \newblock Semi-{{Supervised Multi}}-{{Label Deep Learning}} Based
1316 | {{Non}}-Intrusive {{Load Monitoring}} in {{Smart Grids}}.
1317 | \newblock {\em IEEE Transactions on Industrial Informatics} {\bf 2019}, pp.
1318 | 1--1.
1319 | \newblock
1320 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TII.2019.2955470}{\detokenize{10.1109/TII.2019.2955470}}}.
1321 |
1322 | \bibitem[Xia \em{et~al.}(2019{\natexlab{a}})Xia, Liu, Wang, Zhang, and
1323 | Xu]{xia2019a}
1324 | Xia, M.; Liu, W.; Wang, K.; Zhang, X.; Xu, Y.
1325 | \newblock Non-Intrusive Load Disaggregation Based on Deep Dilated Residual
1326 | Network.
1327 | \newblock {\em Electric Power Systems Research} {\bf 2019}, {\em
1328 | 170},~277--285.
1329 | \newblock
1330 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1016/j.epsr.2019.01.034}{\detokenize{10.1016/j.epsr.2019.01.034}}}.
1331 |
1332 | \bibitem[Xia \em{et~al.}(2019{\natexlab{b}})Xia, Liu, Xu, Wang, and
1333 | Zhang]{xia2019}
1334 | Xia, M.; Liu, W.; Xu, Y.; Wang, K.; Zhang, X.
1335 | \newblock Dilated Residual Attention Network for Load Disaggregation.
1336 | \newblock {\em Neural Computing and Applications} {\bf 2019}.
1337 | \newblock
1338 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/s00521-019-04414-3}{\detokenize{10.1007/s00521-019-04414-3}}}.
1339 |
1340 | \bibitem[Yu \em{et~al.}(2019)Yu, Jiang, Li, Zhou, Wang, Cheng, and Gu]{yu2019}
1341 | Yu, H.; Jiang, Z.; Li, Y.; Zhou, J.; Wang, K.; Cheng, Z.; Gu, Q.
1342 | \newblock A {{Multi}}-Objective {{Non}}-Intrusive {{Load Monitoring Method
1343 | Based}} on {{Deep Learning}}.
1344 | \newblock {\em IOP Conference Series: Materials Science and Engineering} {\bf
1345 | 2019}, {\em 486},~012110.
1346 | \newblock
1347 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1088/1757-899X/486/1/012110}{\detokenize{10.1088/1757-899X/486/1/012110}}}.
1348 |
1349 | \bibitem[Popa \em{et~al.}(2019)Popa, Pop, Serbanescu, and
1350 | Castiglione]{popa2019}
1351 | Popa, D.; Pop, F.; Serbanescu, C.; Castiglione, A.
1352 | \newblock Deep Learning Model for Home Automation and Energy Reduction in a
1353 | Smart Home Environment Platform.
1354 | \newblock {\em Neural Computing and Applications} {\bf 2019}, {\em
1355 | 31},~1317--1337.
1356 | \newblock
1357 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/s00521-018-3724-6}{\detokenize{10.1007/s00521-018-3724-6}}}.
1358 |
1359 | \bibitem[Wang \em{et~al.}(2019)Wang, Ji, and Li]{wang2019a}
1360 | Wang, T.S.; Ji, T.Y.; Li, M.S.
1361 | \newblock A {{New Approach}} for {{Supervised Power Disaggregation}} by
1362 | {{Using}} a {{Denoising Autoencoder}} and {{Recurrent LSTM Network}}.
1363 | \newblock 2019 {{IEEE}} 12th {{International Symposium}} on {{Diagnostics}}
1364 | for {{Electrical Machines}}, {{Power Electronics}} and {{Drives}}
1365 | ({{SDEMPED}}); {IEEE}: {Toulouse, France}, 2019; pp. 507--512.
1366 | \newblock
1367 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/DEMPED.2019.8864870}{\detokenize{10.1109/DEMPED.2019.8864870}}}.
1368 |
1369 | \bibitem[Fagiani \em{et~al.}(2019)Fagiani, Bonfigli, Principi, Squartini, and
1370 | Mandolini]{fagiani2019}
1371 | Fagiani, M.; Bonfigli, R.; Principi, E.; Squartini, S.; Mandolini, L.
1372 | \newblock A {{Non}}-{{Intrusive Load Monitoring Algorithm Based}} on
1373 | {{Non}}-{{Uniform Sampling}} of {{Power Data}} and {{Deep Neural Networks}}.
1374 | \newblock {\em Energies} {\bf 2019}, {\em 12},~1371.
1375 | \newblock
1376 | doi:{\changeurlcolor{black}\href{https://doi.org/10.3390/en12071371}{\detokenize{10.3390/en12071371}}}.
1377 |
1378 | \bibitem[Bejarano \em{et~al.}(2019)Bejarano, DeFazio, and Ramesh]{bejarano2019}
1379 | Bejarano, G.; DeFazio, D.; Ramesh, A.
1380 | \newblock Deep {{Latent Generative Models For Energy Disaggregation}}.
1381 | \newblock Proceedings of the {{AAAI Conference}} on {{Artificial
1382 | Intelligence}}; , 2019; Vol.~33.
1383 | \newblock
1384 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1609/aaai.v33i01.3301850}{\detokenize{10.1609/aaai.v33i01.3301850}}}.
1385 |
1386 | \bibitem[Cho \em{et~al.}(2018)Cho, Hu, and Sartipi]{cho2018}
1387 | Cho, J.; Hu, Z.; Sartipi, M.
1388 | \newblock Non-{{Intrusive A}}/{{C Load Disaggregation Using Deep Learning}}.
1389 | \newblock 2018 {{IEEE}}/{{PES Transmission}} and {{Distribution Conference}}
1390 | and {{Exposition}} ({{T D}}), 2018, pp. 1--5.
1391 | \newblock
1392 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TDC.2018.8440358}{\detokenize{10.1109/TDC.2018.8440358}}}.
1393 |
1394 | \bibitem[Van~Zaen \em{et~al.}(2018)Van~Zaen, El~Achkar, Carrillo, and
1395 | Hutter]{vanzaen2018}
1396 | Van~Zaen, J.; El~Achkar, C.M.; Carrillo, R.E.; Hutter, A.
1397 | \newblock Detection and {{Classification}} of {{Refrigeration Units}} in a
1398 | {{Commercial Environment}}: {{Comparing Neural Networks}} to {{Unsupervised
1399 | Clustering}}.
1400 | \newblock 4th {{International Workshop}} on {{Non}}-{{Intrusive Load
1401 | Monitoring}}; {Nilmworkshop Org}: {Austin, Texas}, 2018.
1402 |
1403 | \bibitem[Chang \em{et~al.}(2018)Chang, Chen, and Lin]{chang2018}
1404 | Chang, F.Y.; Chen, C.; Lin, S.D.
1405 | \newblock An {{Empirical Study}} of {{Ladder Network}} and {{Multitask
1406 | Learning}} on {{Energy Disaggregation}} in {{Taiwan}}.
1407 | \newblock 2018 {{Conference}} on {{Technologies}} and {{Applications}} of
1408 | {{Artificial Intelligence}} ({{TAAI}}); {IEEE}: {Taichung}, 2018; pp.
1409 | 86--89.
1410 | \newblock
1411 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TAAI.2018.00028}{\detokenize{10.1109/TAAI.2018.00028}}}.
1412 |
1413 | \bibitem[Sirojan \em{et~al.}(2018)Sirojan, Phung, and
1414 | Ambikairajah]{sirojan2018}
1415 | Sirojan, T.; Phung, B.T.; Ambikairajah, E.
1416 | \newblock Deep {{Neural Network Based Energy Disaggregation}}.
1417 | \newblock 2018 {{IEEE International Conference}} on {{Smart Energy Grid
1418 | Engineering}} ({{SEGE}}), 2018, pp. 73--77.
1419 | \newblock
1420 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SEGE.2018.8499441}{\detokenize{10.1109/SEGE.2018.8499441}}}.
1421 |
1422 | \bibitem[Rafiq \em{et~al.}(2018)Rafiq, Zhang, Li, and Ochani]{rafiq2018}
1423 | Rafiq, H.; Zhang, H.; Li, H.; Ochani, M.K.
1424 | \newblock Regularized {{LSTM Based Deep Learning Model}}: {{First Step}}
1425 | towards {{Real}}-{{Time Non}}-{{Intrusive Load Monitoring}}.
1426 | \newblock 2018 {{IEEE International Conference}} on {{Smart Energy Grid
1427 | Engineering}} ({{SEGE}}), 2018, pp. 234--239.
1428 | \newblock
1429 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SEGE.2018.8499519}{\detokenize{10.1109/SEGE.2018.8499519}}}.
1430 |
1431 | \bibitem[Martins \em{et~al.}(2018)Martins, Gomes, Nascimento, and {de
1432 | Freitas}]{martins2018}
1433 | Martins, P.B.M.; Gomes, J.G.R.C.; Nascimento, V.B.; {de Freitas}, A.R.
1434 | \newblock Application of a {{Deep Learning Generative Model}} to {{Load
1435 | Disaggregation}} for {{Industrial Machinery Power Consumption Monitoring}}.
1436 | \newblock 2018 {{IEEE International Conference}} on {{Communications}},
1437 | {{Control}}, and {{Computing Technologies}} for {{Smart Grids}}
1438 | ({{SmartGridComm}}); {IEEE}: {Aalborg}, 2018; pp. 1--6.
1439 | \newblock
1440 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/SmartGridComm.2018.8587415}{\detokenize{10.1109/SmartGridComm.2018.8587415}}}.
1441 |
1442 | \bibitem[Chen \em{et~al.}(2018)Chen, Wang, He, Chen, Hu, and He]{chen2018c}
1443 | Chen, K.; Wang, Q.; He, Z.; Chen, K.; Hu, J.; He, J.
1444 | \newblock Convolutional Sequence to Sequence Non-Intrusive Load Monitoring.
1445 | \newblock {\em The Journal of Engineering} {\bf 2018}, {\em 2018},~1860--1864.
1446 | \newblock
1447 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1049/joe.2018.8352}{\detokenize{10.1049/joe.2018.8352}}}.
1448 |
1449 | \bibitem[Krystalakos \em{et~al.}(2018)Krystalakos, Nalmpantis, and
1450 | Vrakas]{krystalakos2018}
1451 | Krystalakos, O.; Nalmpantis, C.; Vrakas, D.
1452 | \newblock Sliding {{Window Approach}} for {{Online Energy Disaggregation Using
1453 | Artificial Neural Networks}}.
1454 | \newblock Proceedings of the 10th {{Hellenic Conference}} on {{Artificial
1455 | Intelligence}}; {ACM}: {New York, NY, USA}, 2018; {{SETN}} '18, pp.
1456 | 7:1--7:6.
1457 | \newblock
1458 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/3200947.3201011}{\detokenize{10.1145/3200947.3201011}}}.
1459 |
1460 | \bibitem[Tsai \em{et~al.}(2018)Tsai, Yang, Ho, Yin, and Chiang]{tsai2018}
1461 | Tsai, C.W.; Yang, C.W.; Ho, W.J.; Yin, Z.X.; Chiang, K.C.
1462 | \newblock Using Autoencoder Network to Implement Non-Intrusive Load Monitoring
1463 | of Small and Medium Business Customer.
1464 | \newblock 2018 {{IEEE International Conference}} on {{Applied System
1465 | Invention}} ({{ICASI}}), 2018, pp. 433--436.
1466 | \newblock
1467 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICASI.2018.8394277}{\detokenize{10.1109/ICASI.2018.8394277}}}.
1468 |
1469 | \bibitem[Le \em{et~al.}(2016)Le, Kim, and Kim]{le2016}
1470 | Le, T.T.H.; Kim, J.; Kim, H.
1471 | \newblock Classification Performance Using Gated Recurrent Unit Recurrent
1472 | Neural Network on Energy Disaggregation.
1473 | \newblock 2016 {{International Conference}} on {{Machine Learning}} and
1474 | {{Cybernetics}} ({{ICMLC}}). {IEEE}, 2016, Vol.~1, pp. 105--110.
1475 | \newblock
1476 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/ICMLC.2016.7860885}{\detokenize{10.1109/ICMLC.2016.7860885}}}.
1477 |
1478 | \bibitem[He and Chai(2016)]{he2016}
1479 | He, W.; Chai, Y.
1480 | \newblock An {{Empirical Study}} on {{Energy Disaggregation}} via {{Deep
1481 | Learning}}.
1482 | \newblock 2016 2nd {{International Conference}} on {{Artificial Intelligence}}
1483 | and {{Industrial Engineering}} ({{AIIE}} 2016). {Atlantis Press}, 2016, Vol.
1484 | 133, pp. 338--342.
1485 | \newblock
1486 | doi:{\changeurlcolor{black}\href{https://doi.org/10.2991/aiie-16.2016.77}{\detokenize{10.2991/aiie-16.2016.77}}}.
1487 |
1488 | \bibitem[Ronneberger \em{et~al.}(2015)Ronneberger, Fischer, and
1489 | Brox]{ronneberger2015}
1490 | Ronneberger, O.; Fischer, P.; Brox, T.
1491 | \newblock U-{{Net}}: {{Convolutional Networks}} for {{Biomedical Image
1492 | Segmentation}}. In {\em Medical {{Image Computing}} and
1493 | {{Computer}}-{{Assisted Intervention}} \textendash{} {{MICCAI}} 2015}; Navab,
1494 | N.; Hornegger, J.; Wells, W.M.; Frangi, A.F., Eds.; {Springer International
1495 | Publishing}: {Cham}, 2015; Vol. 9351, pp. 234--241.
1496 | \newblock
1497 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1007/978-3-319-24574-4_28}{\detokenize{10.1007/978-3-319-24574-4_28}}}.
1498 |
1499 | \bibitem[zot()]{zotero-8847}
1500 | {{TensorFlow Hub}} - {{A}} Repository of Trained Machine Learning Models
1501 | (Www.Tensorflow.Org/Hub).
1502 |
1503 | \bibitem[202(2020)]{2020c}
1504 | 14th {{ACM International Conference}} on {{Distributed}} and {{Event}}-{{Based
1505 | Systems DEBS}} 2020 - {{Challenge}}
1506 | (2020.Debs.Org/Call-for-Grand-Challenge-Solutions/), 2020.
1507 |
1508 | \bibitem[Ruder(2017)]{ruder2017}
1509 | Ruder, S.
1510 | \newblock An {{Overview}} of {{Multi}}-{{Task Learning}} in {{Deep Neural
1511 | Networks}}.
1512 | \newblock {\em arXiv:1706.05098 [cs, stat]} {\bf 2017},
1513 | \href{http://xxx.lanl.gov/abs/1706.05098}{{\normalfont [arXiv:cs,
1514 | stat/1706.05098]}}.
1515 |
1516 | \bibitem[Doukas(2019)]{doukas2019}
1517 | Doukas, D.
1518 | \newblock {{EU NILM Workshop}} 2019 - {{NILM}} Datasets, Benchmarking and
1519 | Evaluation
1520 | (Www.Youtube.Com/Watch?V={{v5XoLtQH9Uw}}\&list={{PLJrF}}-{{gxa0ImryGeNtil}}-{{s9zPJOaV4w}}-{{Vy}}\&index=22\&t=0s),
1521 | 2019.
1522 |
1523 | \bibitem[Ouali \em{et~al.}(2020)Ouali, Hudelot, and Tami]{ouali2020}
1524 | Ouali, Y.; Hudelot, C.; Tami, M.
1525 | \newblock An {{Overview}} of {{Deep Semi}}-{{Supervised Learning}}.
1526 | \newblock {\em arXiv:2006.05278 [cs, stat]} {\bf 2020},
1527 | \href{http://xxx.lanl.gov/abs/2006.05278}{{\normalfont [arXiv:cs,
1528 | stat/2006.05278]}}.
1529 |
1530 | \bibitem[Rasmus \em{et~al.}(2015)Rasmus, Valpola, Honkala, Berglund, and
1531 | Raiko]{rasmus2015}
1532 | Rasmus, A.; Valpola, H.; Honkala, M.; Berglund, M.; Raiko, T.
1533 | \newblock Semi-{{Supervised Learning}} with {{Ladder Networks}}.
1534 | \newblock {\em arXiv:1507.02672 [cs, stat]} {\bf 2015},
1535 | \href{http://xxx.lanl.gov/abs/1507.02672}{{\normalfont [arXiv:cs,
1536 | stat/1507.02672]}}.
1537 |
1538 | \bibitem[Tarvainen and Valpola(2017)]{tarvainen2017}
1539 | Tarvainen, A.; Valpola, H.
1540 | \newblock Mean Teachers Are Better Role Models: {{Weight}}-Averaged Consistency
1541 | Targets Improve Semi-Supervised Deep Learning Results.
1542 | \newblock Proceedings of the 31st {{International Conference}} on {{Neural
1543 | Information Processing Systems}}; {Curran Associates Inc.}: {Red Hook, NY,
1544 | USA}, 2017; {{NIPS}}'17, pp. 1195--1204.
1545 |
1546 | \bibitem[Miyato \em{et~al.}(2018)Miyato, Maeda, Koyama, and Ishii]{miyato2018}
1547 | Miyato, T.; Maeda, S.i.; Koyama, M.; Ishii, S.
1548 | \newblock Virtual Adversarial Training: A Regularization Method for Supervised
1549 | and Semi-Supervised Learning.
1550 | \newblock {\em IEEE Transactions on Pattern Analysis and Machine Intelligence}
1551 | {\bf 2018}, {\em 41},~1979--1993.
1552 | \newblock
1553 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1109/TPAMI.2018.2858821}{\detokenize{10.1109/TPAMI.2018.2858821}}}.
1554 |
1555 | \bibitem[Kuo \em{et~al.}(2020)Kuo, Ma, Huang, and Kira]{kuo2020}
1556 | Kuo, C.W.; Ma, C.Y.; Huang, J.B.; Kira, Z.
1557 | \newblock {{FeatMatch}}: {{Feature}}-{{Based Augmentation}} for
1558 | {{Semi}}-{{Supervised Learning}}.
1559 | \newblock {\em arXiv:2007.08505 [cs]} {\bf 2020},
1560 | \href{http://xxx.lanl.gov/abs/2007.08505}{{\normalfont
1561 | [arXiv:cs/2007.08505]}}.
1562 |
1563 | \bibitem[Xie \em{et~al.}(2019)Xie, Dai, Hovy, Luong, and Le]{xie2019}
1564 | Xie, Q.; Dai, Z.; Hovy, E.; Luong, M.T.; Le, Q.V.
1565 | \newblock Unsupervised {{Data Augmentation}} for {{Consistency Training}}.
1566 | \newblock {\em arXiv:1904.12848 [cs, stat]} {\bf 2019},
1567 | \href{http://xxx.lanl.gov/abs/1904.12848}{{\normalfont [arXiv:cs,
1568 | stat/1904.12848]}}.
1569 |
1570 | \bibitem[French \em{et~al.}(2020)French, Oliver, and Salimans]{french2020}
1571 | French, G.; Oliver, A.; Salimans, T.
1572 | \newblock Milking {{CowMask}} for {{Semi}}-{{Supervised Image Classification}}.
1573 | \newblock {\em arXiv:2003.12022 [cs]} {\bf 2020},
1574 | \href{http://xxx.lanl.gov/abs/2003.12022}{{\normalfont
1575 | [arXiv:cs/2003.12022]}}.
1576 |
1577 | \bibitem[Howard \em{et~al.}(2017)Howard, Zhu, Chen, Kalenichenko, Wang, Weyand,
1578 | Andreetto, and Adam]{howard2017}
1579 | Howard, A.G.; Zhu, M.; Chen, B.; Kalenichenko, D.; Wang, W.; Weyand, T.;
1580 | Andreetto, M.; Adam, H.
1581 | \newblock {{MobileNets}}: {{Efficient Convolutional Neural Networks}} for
1582 | {{Mobile Vision Applications}}.
1583 | \newblock {\em arXiv:1704.04861 [cs]} {\bf 2017},
1584 | \href{http://xxx.lanl.gov/abs/1704.04861}{{\normalfont
1585 | [arXiv:cs/1704.04861]}}.
1586 |
1587 | \bibitem[Howard \em{et~al.}(2018)Howard, Zhmoginov, Chen, Sandler, and
1588 | Zhu]{howard2018}
1589 | Howard, A.; Zhmoginov, A.; Chen, L.C.; Sandler, M.; Zhu, M.
1590 | \newblock Inverted Residuals and Linear Bottlenecks: {{Mobile}} Networks for
1591 | Classification, Detection and Segmentation.
1592 | \newblock {{CVPR}} 2018; , 2018.
1593 |
1594 | \bibitem[Sze(2019)]{sze2019}
1595 | Sze, V.
1596 | \newblock Efficient {{Processing}} of {{Deep Neural Networks}}: From
1597 | {{Algorithms}} to {{Hardware Architectures}}.
1598 | \newblock {{NeurIPS}} 2020; , 2019.
1599 |
1600 | \bibitem[Batra \em{et~al.}(2013)Batra, Gulati, Singh, and
1601 | Srivastava]{batra2013}
1602 | Batra, N.; Gulati, M.; Singh, A.; Srivastava, M.B.
1603 | \newblock It's {{Different}}: {{Insights}} into Home Energy Consumption in
1604 | {{India}}.
1605 | \newblock Proceedings of the 5th {{ACM Workshop}} on {{Embedded Systems For
1606 | Energy}}-{{Efficient Buildings}}; {Association for Computing Machinery}: {New
1607 | York, NY, USA}, 2013; {{BuildSys}}'13, pp. 1--8.
1608 | \newblock
1609 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1145/2528282.2528293}{\detokenize{10.1145/2528282.2528293}}}.
1610 |
1611 | \bibitem[Kriechbaumer and Jacobsen(2018)]{kriechbaumer2018}
1612 | Kriechbaumer, T.; Jacobsen, H.A.
1613 | \newblock {{BLOND}}, a Building-Level Office Environment Dataset of Typical
1614 | Electrical Appliances.
1615 | \newblock {\em Scientific Data} {\bf 2018}, {\em 5},~1--14.
1616 | \newblock
1617 | doi:{\changeurlcolor{black}\href{https://doi.org/10.1038/sdata.2018.48}{\detokenize{10.1038/sdata.2018.48}}}.
1618 |
1619 | \bibitem[Cohen and Welling(2016)]{cohen2016}
1620 | Cohen, T.; Welling, M.
1621 | \newblock Group Equivariant Convolutional Networks.
1622 | \newblock Proceedings of {{The}} 33rd {{International Conference}} on
1623 | {{Machine Learning}}; {PMLR}: {New York, New York, USA}, 2016; Vol.~48, {\em
1624 | \{\vphantom\}{{Proceedings}} of {{Machine Learning Research}}}, pp.
1625 | 2990--2999.
1626 |
1627 | \bibitem[Zaheer \em{et~al.}(2017)Zaheer, Kottur, Ravanbakhsh, Poczos,
1628 | Salakhutdinov, and Smola]{zaheer2017}
1629 | Zaheer, M.; Kottur, S.; Ravanbakhsh, S.; Poczos, B.; Salakhutdinov, R.R.;
1630 | Smola, A.J.
1631 | \newblock Deep Sets.
1632 | \newblock Advances in Neural Information Processing Systems; {Curran
1633 | Associates, Inc.}: {Long Beach, CA, USA}, 2017; Vol.~30, pp. 3391--3401.
1634 |
1635 | \bibitem[McMahan \em{et~al.}(2017)McMahan, Moore, Ramage, Hampson, and
1636 | y~Arcas]{mcmahan2017}
1637 | McMahan, B.; Moore, E.; Ramage, D.; Hampson, S.; y~Arcas, B.A.
1638 | \newblock Communication-{{Efficient Learning}} of {{Deep Networks}} from
1639 | {{Decentralized Data}}.
1640 | \newblock Proceedings of the 20th {{International Conference}} on {{Artificial
1641 | Intelligence}} and {{Statistics}}; {PMLR}: {Fort Lauderdale, Florida, USA},
1642 | 2017; Vol.~54, pp. 1273--1282.
1643 |
1644 | \bibitem[Kairouz \em{et~al.}(2019)Kairouz, McMahan, Avent, Bellet, Bennis,
1645 | Bhagoji, Bonawitz, Charles, Cormode, Cummings, D'Oliveira, Rouayheb, Evans,
1646 | Gardner, Garrett, Gasc{\'o}n, Ghazi, Gibbons, Gruteser, Harchaoui, He, He,
1647 | Huo, Hutchinson, Hsu, Jaggi, Javidi, Joshi, Khodak, Kone{\v c}n{\'y},
1648 | Korolova, Koushanfar, Koyejo, Lepoint, Liu, Mittal, Mohri, Nock,
1649 | {\"O}zg{\"u}r, Pagh, Raykova, Qi, Ramage, Raskar, Song, Song, Stich, Sun,
1650 | Suresh, Tram{\`e}r, Vepakomma, Wang, Xiong, Xu, Yang, Yu, Yu, and
1651 | Zhao]{kairouz2019}
1652 | Kairouz, P.; McMahan, H.B.; Avent, B.; Bellet, A.; Bennis, M.; Bhagoji, A.N.;
1653 | Bonawitz, K.; Charles, Z.; Cormode, G.; Cummings, R.; D'Oliveira, R.G.L.;
1654 | Rouayheb, S.E.; Evans, D.; Gardner, J.; Garrett, Z.; Gasc{\'o}n, A.; Ghazi,
1655 | B.; Gibbons, P.B.; Gruteser, M.; Harchaoui, Z.; He, C.; He, L.; Huo, Z.;
1656 | Hutchinson, B.; Hsu, J.; Jaggi, M.; Javidi, T.; Joshi, G.; Khodak, M.;
1657 | Kone{\v c}n{\'y}, J.; Korolova, A.; Koushanfar, F.; Koyejo, S.; Lepoint, T.;
1658 | Liu, Y.; Mittal, P.; Mohri, M.; Nock, R.; {\"O}zg{\"u}r, A.; Pagh, R.;
1659 | Raykova, M.; Qi, H.; Ramage, D.; Raskar, R.; Song, D.; Song, W.; Stich, S.U.;
1660 | Sun, Z.; Suresh, A.T.; Tram{\`e}r, F.; Vepakomma, P.; Wang, J.; Xiong, L.;
1661 | Xu, Z.; Yang, Q.; Yu, F.X.; Yu, H.; Zhao, S.
1662 | \newblock Advances and {{Open Problems}} in {{Federated Learning}}.
1663 | \newblock {\em arXiv:1912.04977 [cs, stat]} {\bf 2019},
1664 | \href{http://xxx.lanl.gov/abs/1912.04977}{{\normalfont [arXiv:cs,
1665 | stat/1912.04977]}}.
1666 |
1667 | \end{thebibliography}
1668 |
--------------------------------------------------------------------------------