CS229 : Machine Learning
The "ML" course at Stanford , or to say the most popular Machine Learning course Worldwide is CS229. CS229 is Math Heavy and is
No Text , but Pattern Classification - Richard Duda, Peter Hart and David Stork | handout | Syllabus
π Important Books :
π Hands on Machine Learning with Scikit Learn and TensorFlow
π Introduction to Machine Learning - Ethem AlpaydΔ±n
π Machine Learning A Probabilistic Perspective
π Optimization for Machine Learning
π Pattern Recognition and Machine Learning - Bishop
π Pattern Recognition and Machine Learning Solution
Homework (40%) + Mid-term (20%) + Final Project (40%)
π₯€ Homeworks (Problem Sets): Solutions
π¦ PSET 0
π¦ PSET 1 - inst - testdata
π¦ PSET 2 - inst
π¦ PSET 3 - inst - peppers_numpy
π¦ PSET 4 - inst - autograder&solution
Course:
This course provides a broad introduction to machine learning and statistical pattern recognition. Topics include: supervised learning (generative/discriminative learning, parametric/non-parametric learning, neural networks, support vector machines); unsupervised learning (clustering, dimensionality reduction, kernel methods); learning theory (bias/variance tradeoffs; VC theory; large margins); reinforcement learning and adaptive control. The course will also discuss recent applications of machine learning, such as to robotic control, data mining, autonomous navigation, bioinformatics, speech recognition, and text and web data processing.
β INTRODUCTION
π½ Basic concepts.
β SUPERVISED LEARNING
π½ Supervised learning setup. LMS. Supervised Learning, Discriminative Algorithms
π½ Logistic regression. Perceptron. Exponential family. Linear Algebra
π½ Generative learning algorithms. Gaussian discriminant analysis. Naive Bayes.
π½ Support Vector Machines
π½ Model selection and feature selection, Probability
π½ Evaluating and debugging learning algorithms, Python
β LEARNING THEORY
π½ Advice on applying machine learning, Bias/variance tradeoff and error analysis
π½ Regularization and Model Selection
π½ Practical advice on how to use learning algorithms.
β DEEP LEARNING
π½ Neural Networks, Deep Learning, Backpropagation, Trees
π½ Online Learning and the Perceptron Algorithm, Vectorization.
β UNSUPERVISED LEARNING
π½ Clustering. Unsupervised Learning, k-means clustering
π½ EM Algorithm. Mixture of Gaussians
π½ Factor analysis
π½ PCA (Principal components analysis)
π½ ICA (Independent components analysis)
β REINFORCEMENT LEARNING AND CONTROL
π½ MDPs. Bellman equations, Reinforcement Learning and Control
π½ Value iteration and policy iteration. Linear quadratic regulation (LQR), LQR, DDP and LQG
π½ Q-learning. Value function approximation. Policy search. Reinforce. POMDPs.
π½ On critiques of Machine Learning
Supplementary Notes
β Binary classification with +/-1 labels pdf
β Boosting algorithms and weak learning pdf
β Functional after implementing stump_booster.m in PS2. here
β The representer theorem pdf
β Hoeffding's inequality pdf
Section Notes
β Linear Algebra Review and Reference pdf
β Probability Theory Review pdf
β Convex Optimization Overview, Part I pdf
β Convex Optimization Overview, Part II pdf
β Hidden Markov Models pdf
β The Multivariate Gaussian Distribution pdf
β More on Gaussian Distribution pdf
β Gaussian Processes pdf
Exams - 2018 Autumn,
β· My Solutions for CS229 Exams - 2018, 2017, 2016
βΏ MIT ML | ML Cheatsheet | CS229 Cheatsheet | Google Scholar | arXiv | UCI ML dataset repository | CS229#Notes
βΏ Official CS229 Notes | CS229 Notes by Shervine | Machine learning study guides tailored to CS 229 | Derivatives Backpropagation and Vectorization | CS229 More Notes - Section Notes - Supplementary Notes | CS229@2008 - syllabus | Statistical Machine Learning CMU Spring 2016 | MIT Course 9.520 - Statistical Learning Theory and Applications, Fall 2015 | Advanced Introduction to Machine Learning - CMU - Fall 2015
βΏ Cheatsheets :
π― Supervised Learning
π― Unsupervised Learning
π― Machine Learning
π― Deep Learning
π― Algebra Calculus
π― Probability Statistics
π― Combined
π― Machine Learning
π― ML tools
FINAL PROJECT | Past Projects
CS229 gives a lot of importance to Final Project, going through the past projects , I was happy to see the dynamic range of wonderful ideas and application of ML all the way (I have the list - e.g, eg2). Good CS229 projects are either publishable or minor changes to be able to publish the project. NIPS (NeurIPS) , ICML are the ML conferences to show ML works to other people around the world. I was ecstatic to start my own and did " ".