# CS189 Intro to Machine Learning (Schewchuk) - 2022 Spring

Those are note archives as well as very well-structured final review docs I made for my self. I put it here so that you can study it and also I will be able to review it later. So if you found this useful, make sure you star my repo

Additional Note: The credit of most of those notes are due to the instructor (Prof. Johnathan Shewchuk) and TAs of this course. A great thank you for them for making this course so enjoyable!

## Resource List

- Resources
- Notes (Credit to Prof. Shewchuk)
- Lec 01 - Introduction, Overview and Logistics
- Lec 02 - Linear Classifier (Perceptrons)
- Lec 03 - Perceptron Continued, Gradient Descent, weight space transform and SVM
- Lec 04 - Soft Margin SVM, Feature Lifting
- Lec 05 - ML Abstractions / Optimization(Linear/Quadratic Program)
- Lec 06 - Decision Theory / Risk Minimization Technique
- Lec 07 - Gaussian Discriminant Analysis (GDA => LDA/GDA) with Isotropic Gaussian
- Lec 08 - Eigenvectors / Quadratic Forms / GDA with Anisotropic Gaussian
- Lec 09 - More GDA with Anisotropic Gaussian / Data Cleaning Vocabs
- Lec 10 - Regression(Intro/Logistic)
- Lec 11 - Regression(Least-Squre Polynomial / Weighted Least Square / Logistic), Newton’s Method, ROC Curves
- Lec 12 - Statistical Justification for Least Square / Empirical Risk / Logistic Loss, Bias-Variance Decomposition
- Lec 13 - Shrinkage(Ridge/Lasso Regression, Feature Selection)
- Lec 14 - Decision Trees
- Lec 15 - Multivariate Decision Trees, Decision Tree Regression, Pruning, Ensemble Learning(Bagging, Random Forest)
- Lec 16 - Kernel Trick (example on Ridge Regression / Perceptron / Logistic Regression), RBF and Polynomial Kernel
- Lec 17 - Neural Nets
- Lec 18 - Neurobiology and Variations on Neural Networks
- Lec 19 - Better NN Training, CNN
- Lec 20 - Unsupervised Learning: Principle Component Analysis (PCA)
- Lec 21 - SVD & Clustering
- Lec 22 - High Dimensional Intuition, Random Projection, Pseudoinverse
- Lec 23 - Learning Theory - Shatter Function / VC Dimension
- Lec 24 - Adaboost & k-Nearest Neighbour
- Lec 25 - kNN Algorithms(Exhaustive kNN search, Voronoi Diagrams, k-d trees)

- Alternative Note
- Resources
- Intuition for Decision Boundary | ML Visualizer by Sagnik Bhattacharya, Colin Zhou, Komila Khamidova, and Aaron Sun
- Math Review | Math for ML by Garrett Thomas
- Intuition Bulding for Linear Algebra | 3Blue1Brown - Essence of Linear Algebra
- Matrix Calculus | First two chapters of The Matrix Cookbook
- Stanford CS229 ML Probability Review
- Stanford CS229 Linear Algebra Review
- Vector Derivatives by Stanford CS231N
- CS189 - Everything about Gradients
- Stanford CS231N Linear Backprop Handout
- Counterintuitive Properties of High Dimensional Space
- AdaBoost clearly explained by StatQuest
- Maximum a posteriori (MAP) estimation by mathematicalmonk
- Maximum Likelihood, clearly explained!!! by StatQuest
- What are Maximum Likelihood (ML) and Maximum a posteriori (MAP)? by Iain