LecturesReading and Materials
Week 1 Welcome to SML310 and the Course Projects, K Nearest Neighbors, Linear Regression, generalization and overfitting. Code: Gradient descent. CIML Chapter 3, Chapter 7.
Week 2 Intro to Maximum Likelihood inference
Mini-Project 0 intro
Intro to Bayesian inference, Bayesian inference about unicorns.
Code: Bayesian inference about coins.
CIML Chapter 9. Videos: Unicorns and Bayesian Inference, Why I don't believe in large coefficients, Why L1 regularization drives some coefficients to 0
Week 3 Linear Classifiers CIML Chapter 3, Chapter 7.
Week 4 Generative models and Naive Bayes, Neural Networks CIML Ch. 9. Goodfellow et al.
Week 5 Neural Networks and Generalization Goodfellow et al. Ch. 6
Week 6 Intro to ConvNets, Training Neural Networks.

Intro to PyTorch, Maximum Likelihood with Pytorch, Classifying Digits with Pytorch (mnist_all.mat)

Goodfellow et al., Ch.6-8. Tutorials at
Week 7 Mixture of Gaussians and Unsupervised Learning, generating data from mixtures of Gaussians. Bishop, Ch. 1.2.4, Ch. 2.3, Ch. 9.
Week 8

Unsupervised Learning, Unsupervised Learning II,, laska.png.

Goodfellow et al, Ch. 10. and Ch. 14. Andrej Karpathy, The Unreasonable Effectiveness of Recurrent Neural Networks

Studies mentioned/discussed in class:

Week 9 Statistical Inference

finches.Rms, finches.html.

Shalizi, Chapters 1-3.

Studies mentioned/discussed in class:

Week 10 "The Truth about Linear Regression"

Intro to Hierarchical Models (R code: multi.html, multi.Rmd, srrs2.dat)

Hierarchical models: polls html, Rmd, polls.subset.dat, polls.dta

Case study: hierarchical models and ranking restaraunts by food safety compliance

Causal inference

Shalizi, Chapters 1-3. Shalizi & Gelman, Philosophy and the Practice of Bayesian Inference. Gelman & Hill, Data Analysis Using Regression and Multilevel/Hierarchical Models, Ch. 12.

Shalizi, Chapters 21-24.

Week 11 An intro to Fairness in Machine Leaninrg

The Red Car example: generating fake data and inference with RStan.

The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning. NIPS 2017 tutorial on Fairness in Machine Learning (slides, video. The accuracy, fairness, and limits of predicting recidivism.

Studies mentioned/discussed in class:

Week 12/13 Student presentations

Studies mentioned/discussed in class: