├── 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 | " \n", 818 | " \n", 819 | " \n", 820 | " \n", 821 | " \n", 822 | " \n", 823 | " \n", 824 | " \n", 825 | " \n", 826 | " \n", 827 | " \n", 828 | " \n", 829 | " \n", 830 | " \n", 831 | " \n", 832 | " \n", 833 | " \n", 834 | " \n", 835 | " \n", 836 | " \n", 837 | " \n", 838 | " \n", 839 | " \n", 840 | " \n", 841 | " \n", 842 | " \n", 843 | " \n", 844 | " \n", 845 | " \n", 846 | " \n", 847 | " \n", 848 | " \n", 849 | " \n", 850 | " \n", 851 | " \n", 852 | " \n", 853 | " \n", 854 | " \n", 855 | " \n", 856 | " \n", 857 | " \n", 858 | "
authorscounts
2murray20196
3nascimento20163
0rafiq20202
4cavdar20192
5massidda20202
1sudoso20191
6barsim20181
\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 | --------------------------------------------------------------------------------