I. Introduction to ML Nearest neighbor The ML landscape The statistical learning framework

II. Classification using generative models The generative approach to classification The multivariate Gaussian Exponential families and maximum entropy

III. Linear prediction Linear regression Logistic regression Unconstrained optimization Support vector machines Convex programs and duality Multiclass and structured output prediction

IV. Beyond linear prediction Feedforward neural nets: architecture, backpropagation, convolutional units Kernel methods Decision trees Boosting Random forests

V. Representation learning Clustering Linear projections: PCA and SVD Embeddings and manifold learning Autoencoders and distributed representations

VI. Generalization and confidence Theory of generalization Prediction with calibrated confidence

This repository mainly focuses on five homework assignments assigned during Winter 2019 quarter including theoretical mathematical problems and coding projects. PLEASE strictly follow the Academic Integrity announced in UCSD. Feel free to email me if you have any problems: [email protected].