└── README.md /README.md: -------------------------------------------------------------------------------- 1 | # Some resources for ML research 2 | 3 | Personal and biased selection of ML resources. 4 | 5 | **Disclaimer:** I'm a noivce in ML research, and I read only a few of the list. 6 | 7 | 8 | ## Table of Contents 9 | - [Beginner's Guide](#beginners-guide) 10 | - [Machine Learning](#machine-learning) 11 | - [Deep Learning](#deep-learning) 12 | - [Generative Model](#generative-model) 13 | - [Reinforcement Learning](#reinforcement-learning) 14 | - [Graphical Model](#graphical-model) 15 | - [Optimization](#optimization) 16 | - [Learning Theory](#learning-theory) 17 | - [Statistics](#statistics) 18 | - [Topics in Machine Learning](#topics-in-machine-learning) 19 | - [Math Backgrounds](#math-backgrounds) 20 | - [Blogs](#blogs) 21 | 22 | 23 | ## Beginner's Guide 24 | 25 | **Must Read** 26 | - Machine Learning: A Probabilistic Perspective (Murphy) 27 | - Deep Learning (Goodfellow et al.) 28 | - Reinforcement Learning: An Introduction (Sutton & Barto) 29 | 30 | **Recommended** 31 | - Convex Optimization (Boyd & Vandenberghe) 32 | - Graphical Models, Exponential Families, and Variational Inference (Wainwright & Jordan) 33 | - Learning from Data (Abu-Mostafa) *-> for whom interested in learning theory* 34 | 35 | **Recent Topics** 36 | - Read research blogs (e.g., [OpenAI](https://blog.openai.com/), [BAIR](http://bair.berkeley.edu/blog/), [CMU](https://blog.ml.cmu.edu/)) 37 | - Read lectures from Berkeley, Stanford, CMU or UofT (e.g., [unsupervised learning](https://sites.google.com/view/berkeley-cs294-158-sp19)) 38 | - There are lots of good sources, but I stop updating them up-to-date 39 | 40 | 41 | ## Machine Learning 42 | There are many ML books, but most of them are encyclopedic.
43 | I recommend to take a course using Murphy or Bishop book. 44 | 45 | ### Textbook 46 | - Machine Learning: A Probabilistic Perspective (Murphy) :sparkles: 47 | - Pattern Recognition and Machine Learning (Bishop) :sparkles: 48 | - The Elements of Statistical Learning (Hastie et al.) 49 | - Pattern Classification (Duda et al.) 50 | - Bayesian Reasoning and Machine Learning (Barber) 51 | 52 | ### Lecture 53 | - [Stanford CS229 Machine Learning](http://cs229.stanford.edu) :sparkles: 54 | - [CMU 10701 Mahine Learning](http://www.cs.cmu.edu/~tom/10701_sp11/) 55 | - [CMU 10702 Statistical Machine Learning](http://www.stat.cmu.edu/~larry/=sml/) 56 | - [Oxford Machine Learning](https://www.cs.ox.ac.uk/people/nando.defreitas/machinelearning/) 57 | 58 | 59 | ## Deep Learning 60 | Goodfellow et al. is the new classic.
61 | For vision and NLP, Stanford lectures would be helpful. 62 | 63 | ### Textbook 64 | - Deep Learning (Goodfellow et al.) :sparkles: 65 | 66 | ### Lecture (Practice) 67 | - [Deep Learning book](http://www.deeplearningbook.org/lecture_slides.html) :sparkles: 68 | - [Stanford CS231n Convolutional Neural Networks for Visual Recognition](http://cs231n.stanford.edu) :sparkles: 69 | - [Stanfrod CS224d Deep Learning for Natural Language Processing](http://cs224d.stanford.edu) 70 | - [Stanfrod CS224s Spoken Language Processing](http://web.stanford.edu/class/cs224s/) 71 | - [Oxford Deep Natural Language Processing](https://github.com/oxford-cs-deepnlp-2017/lectures) 72 | - [CMU 11747 Neural Networks for NLP](http://phontron.com/class/nn4nlp2017/index.html) 73 | 74 | ### Lecture (Theory) 75 | - [Stanford Stat385 Theories of Deep Learning](https://stats385.github.io/) 76 | 77 | 78 | ## Generative Model 79 | I seperated generative model as an independent topic,
80 | since I think it is big and important area. 81 | 82 | ### Lecture 83 | - [Toronto CSC2541 Differentiable Inference and Generative Models](https://www.cs.toronto.edu/~duvenaud/courses/csc2541/index.html) 84 | - [Toronto CSC2547 Learning Discrete Latent Structure](https://duvenaud.github.io/learn-discrete/) 85 | - [Toronto CSC2541 Scalable and Flexible Models of Uncertainty](https://csc2541-f17.github.io/) 86 | 87 | 88 | ## Reinforcement Learning 89 | For classic (non-deep) RL, Sutton & Barto is the classic.
90 | For deep RL, lectures from Berkeley/CMU looks good. 91 | 92 | ### Textbook 93 | - Reinforcement Learning: An Introduction (Sutton & Barto) :sparkles: 94 | 95 | ### Lecture 96 | - [UCL Reinforcement Learning](http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching.html) :sparkles: 97 | - [Berkeley CS294 Deep Reinforcement Leanring](http://rll.berkeley.edu/deeprlcourse/) :sparkles: 98 | - [CMU 10703 Deep Reinforcement Learing and Control](https://katefvision.github.io/) 99 | 100 | 101 | ## Graphical Model 102 | Koller & Friedman is comprehensive, but too encyclopedic.
103 | I recommend to take an introductory course using Koller & Friedman book.
104 | 105 | Wainwright & Jordan only focuses on variational inference,
106 | but it gives really good intuition for probabilistic models. 107 | 108 | ### Textbook 109 | - Probabilistic Graphical Models: Principles and Techniques (Koller & Friedman) 110 | - Graphical Models, Exponential Families, and Variational Inference (Wainwright & Jordan) :sparkles: 111 | 112 | ### Lecture 113 | - [Stanford CS228 Probabilistic Graphical Models](http://cs.stanford.edu/~ermon/cs228/index.html) 114 | - [CMU 10708 Probabilistic Graphical Models](http://www.cs.cmu.edu/~epxing/Class/10708/) :sparkles: 115 | 116 | 117 | ## Optimization 118 | Boyd & Vandenberghe is the classic, but I think it's too boring.
119 | Reading chapter 2-5 would be enough. 120 | 121 | Bertsekas more concentrates on convex analysis.
122 | Nocedal & Wright more concentrates on optimization. 123 | 124 | ### Textbook 125 | - Convex Optimization (Boyd & Vandenberghe) :sparkles: 126 | - Convex Optimization Theory (Bertsekas) 127 | - Numerical Optimization (Nocedal & Wright) 128 | 129 | ### Lecture 130 | - [Stanford EE364a Convex Optimization I](http://stanford.edu/class/ee364a/) :sparkles: 131 | - [Stanford EE364b Convex Optimization II](http://stanford.edu/class/ee364a/) 132 | - [MIT 6.253 Convex Analysis and Optimization](https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-253-convex-analysis-and-optimization-spring-2012/index.htm) 133 | 134 | 135 | ## Learning Theory 136 | In my understanding, there are two major topics in learning theory: 137 | 138 | - **Learning Theory:** VC-dimension, PAC-learning 139 | - **Online Learning:** regret bound, multi-armed bandit 140 | 141 | For learning theory, Kearns & Vazirani is the classic; but it's too old-fashined.
142 | Abu-Mostafa is a good introductory book, and I think it's enough for most people. 143 | 144 | For online learning, Cesa-Bianchi & Lugosi is the classic.
145 | For multi-armed bandit, Bubeck & Cesa-Bianchi provides a good survey. 146 | 147 | ### Textbook (Learning Theory) 148 | - Learning from Data (Abu-Mostafa) :sparkles: 149 | - Foundations of Machine Learning (Mohri et al.) 150 | - An Introduction to Computational Learning Theory (Kearns & Vazirani) 151 | 152 | ### Textbook (Online Learning) 153 | - Prediction, Learning, and Games (Cesa-Bianchi & Lugosi) 154 | - Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems (Bubeck & Cesa-Bianchi) 155 | 156 | ### Lecture 157 | - [Caltech Learning from Data](https://work.caltech.edu/telecourse.html) :sparkles: 158 | - [CMU 15859 Machine Learning Theory](http://www.cs.cmu.edu/~avrim/ML14/) 159 | - [Berkeley CS281b/Stat241b Statistical Learning Theory](https://www.stat.berkeley.edu/~bartlett/courses/2014spring-cs281bstat241b/) 160 | - [MIT 9.520 Statistical Learning Theory and Applications](http://www.mit.edu/~9.520/fall15/) 161 | 162 | 163 | ## Statistics 164 | Statistics is a broad area; hence, I listed only a few of them.
165 | For advanced topics, lectures from Berkeley/Stanford/CMU/MIT looks really cool.
166 | 167 | ### Textbook (Statistical Inference) 168 | - All of Statistics (Wasserman) 169 | - Computer Age Statistical Inference (Efron & Hastie) :sparkles: 170 | - Time Series Analysis and Its Applications: With R Examples (Shumway & Stoffer) 171 | 172 | ### Textbook (Nonparametrics) 173 | - All of Nonparametric Statistics (Wasserman) 174 | - Introduction to Nonparametric Estimation (Tsybakov) 175 | - Gaussian Process and Machine Learning (Rasmussen & Williams) :sparkles: 176 | - Bayesian Nonparametrics (Ghosh & Ramamoorthi) :sparkles: 177 | 178 | ### Textbook (Advanced Topics) 179 | - High-Dimensional Statistics: A Non-Asymptotic Viewpoint (Wainwright) :sparkles: 180 | - Statistics for High-Dimensional Data (Bühlmann & van de Geer) 181 | - Asymptotic Statistics (van der Vaart) 182 | - Empirical Processes in M-Estimation (van der Vaart) 183 | 184 | ### Lecture 185 | - [Berkeley Stat210a Theoretical Statistics I](https://www.stat.berkeley.edu/~wfithian/courses/stat210a/) 186 | - [Berkeley Stat210b Theoretical Statistics II](https://people.eecs.berkeley.edu/~jordan/courses/210B-spring17/) 187 | - [Stanford Stat300a Theory of Statistics](https://web.stanford.edu/~lmackey/stats300a/) 188 | - [Stanford CS369m Algorithms for Massive Data Set Analysis](http://cs.stanford.edu/people/mmahoney/cs369m/) 189 | - [CMU 36755 Advanced Statistical Theory I](http://www.stat.cmu.edu/~arinaldo/36755/F16/) 190 | - [MIT 18.S997 High-Dimensional Statistics](https://ocw.mit.edu/courses/mathematics/18-s997-high-dimensional-statistics-spring-2015/) 191 | 192 | 193 | ## Topics in Machine Learning 194 | Miscellaneous topics related to machine learning.
195 | There are much more subfields, but I'll not list them all. 196 | 197 | ### Information Theory 198 | - Elements of Information Theory (Cover & Thomas) 199 | - Information Theory, Inference, and Learning Algorithms (MacKay) 200 | 201 | ### Network Science 202 | - Networks, Crowds, and Markets (Easley & Kleinberg) 203 | - Social and Economic Networks (Jackson) 204 | 205 | ### Markov Chain 206 | - Markov Chains (Norris) 207 | - Markov Chains and Mixing Times (Levin et al.) 208 | 209 | ### Game Theory 210 | - Algorithmic Game Theory (Nisan et al.) 211 | - Multiagent Systems (Shoham & Leyton-Brown) 212 | 213 | ### Combinatorics 214 | - The Probabilistic Method (Alon & Spencer) 215 | - A First Course in Combinatorial Optimization (Lee) 216 | 217 | ### Algorithm 218 | - Introduction to Algorithms (Cormen et al.) 219 | - Randomized Algorithms (Motwani & Raghavan) 220 | - Approximation Algorithms (Vazirani) 221 | 222 | ### Geometric View 223 | - Topological Data Analysis (Wasserman) 224 | - Methods of Information Geometry (Amari & Nagaoka) 225 | - Algebraic Geometry and Statistical Learning Theory (Watanabe) 226 | 227 | ### Some Lectures 228 | - [MIT 18.409 Algorithmic Aspects of Machine Learning](http://people.csail.mit.edu/moitra/409.html) 229 | - [MIT 18.409 An Algorithmist's Toolkit](http://stellar.mit.edu/S/course/18/fa09/18.409/) 230 | 231 | 232 | ## Math Backgrounds 233 | I selected essential topics for machine learning.
234 | Personally, I think more analysis / matrix / geometry never hurts. 235 | 236 | ### Probability 237 | - Probability: Theory and Examples (Durrett) 238 | - Theoretical Statistics (Keener) 239 | - Stochastic Processes (Bass) 240 | - Probability and Statistics Cookbook (Vallentin) 241 | 242 | ### Linear Algebra 243 | - Linear Algebra (Hoffman & Kunze) 244 | - Matrix Analysis (Horn & Johnson) 245 | - Matrix Computations (Golub & Van Loan) 246 | - The Matrix Cookbook (Petersen & Pedersen) 247 | 248 | ### Large Deviations 249 | - Concentration Inequalities and Martingale Inequalities (Chung & Lu) 250 | - An Introduction to Matrix Concentration Inequalities (Tropp) 251 | 252 | 253 | ## Blogs 254 | 255 | - [Google AI Blog](https://ai.googleblog.com/) 256 | - [DeepMind Blog](https://deepmind.com/blog/?category=research) 257 | - [OpenAI Blog](https://blog.openai.com/) 258 | - [FAIR Blog](https://research.fb.com/blog/) 259 | - [Distill.pub](https://distill.pub/) 260 | - [BAIR Blog](http://bair.berkeley.edu/blog/) 261 | - [CMU Blog](https://blog.ml.cmu.edu/) 262 | - [Off the convex path](http://www.offconvex.org/) 263 | - [inFERENCe](http://www.inference.vc/) 264 | - [Sebastian Ruder](http://ruder.io/#open) 265 | - [Lunit Tech Blog (Korean)](https://blog.lunit.io/category/paper-review/) 266 | --------------------------------------------------------------------------------