Overview
Core reference: MIT 6.867 Lecture Notes.
Lecture Schedule & Upload Targets
| # | Topic | Materials |
|---|---|---|
| Lecture 01 | Introduction, linear classification, perceptron update rule |
|
| Lecture 02 | Perceptron convergence, generalization |
|
| Lecture 03 | Maximum margin classification |
|
| Lecture 04 | Classification errors, regularization, logistic regression |
|
| Lecture 05 | Linear regression, estimator bias & variance, active learning |
|
| Lecture 06 | Active learning (cont.), non-linear predictions, kernels |
|
| Lecture 07 | Kernel regression, kernels |
|
| Lecture 08 | Support vector machines (SVMs) & kernels; kernel optimization |
|
| Lecture 09 | Model selection |
|
| Lecture 10 | Model selection criteria |
|
| Lecture 11 | Description length (MDL), feature selection |
|
| Lecture 12 | Combining classifiers, boosting |
|
| Lecture 13 | Boosting, margin, and complexity |
|
| Lecture 14 | Margin & generalization, mixture models |
|
| Lecture 15 | Mixtures and the expectation–maximization (EM) algorithm |
|
| Lecture 16 | EM, regularization, clustering |
|
| Lecture 17 | Clustering |
|
| Lecture 18 | Spectral clustering, Markov models |
|
| Lecture 19 | Hidden Markov models (HMMs) |
|
| Lecture 20 | HMMs (cont.) |
|
| Lecture 21 | Bayesian networks |
|
| Lecture 22 | Learning Bayesian networks |
|
| Lecture 23 | Probabilistic inference; guest lecture: collaborative filtering |
|
| Lecture 24 | Current problems in machine learning; wrap up |
|