Recommended literature

1. Optimal statistical decisions, DeGroot.
2. The Practice of Reproducible Research: Case Studies and Lessons from the Data-Intensive Sciences
3. Privacy, Big Data and the Public Good, Julia Lane et al.
4. The Algorithmic Foundations of Differential Privacy, Dwork and Roth.
5. Decision Making Under Uncertainty and Reinforcement Learning, Dimitrakakis and Ortner

Useful links

• Sign up for Q&A
• Clone the repo
• Videos

Schedule

• 22-23 August: Reproducibility; kNN; Bayesian Inference (Notes: Sec 1, Sec 2.1-2.3. Additional reading: any examples from The Practice of Reproducible Research for background. )
• 29-30 August: Decision problems; Hypothesis Testing; Neural networks; Stochastic gradient descent (Sec. 2.4 - 2.6. Additional reference material: Chapters 6-9 in DeGroot)
• 5-6 September: Tutorial and credit risk lab
• 12-13 September: Databases; k-anonymity; differential privacy. (Notes: Chapter 3, Additional reference material: chapters 1-3.5 in the privacy book, Additional background material: More slides on anonymity and privacy, Privacy, Big Data and the Public Good)
• 19-20 September: Tutorial and privacy lab
• 26-27 September: Fairness; graphical models (Notes: Chapter 4)
• 3-4 October: Tutorial and fairness lab
• 5 October: Project 1 deadline
• 10-11 October: Recommendation Systems; Latent Variables; Importance Sampling (Notes: Chapter 5)
• 17-18 October: Tutorial and recommendation lab
• 24-25 October: Causality, interventions and counterfactuals (Notes: Chapter 6. Additional reference material: Pearl's Causality book. Additional background material: github/papers/causality/)
• 31 Oct-1 Nov: Causality lab
• 7-8 Nov: Bandit problems and experiment design (Notes: Chapter 7. Optionally 8.1, 8.2. Additional reference material: Chapter 6 of Dimitrakakis and Ortner.)
• 14-15 Nov: Experiment design lab
• 21-22 Nov: Project presentations
• 30 Nov: Project 2 deadline