Lecture notes are taken by students. The scribe is expected to verify the correctness of all proofs, to fill in any obvious gaps in the lecture, and to add figures and illustrative images as needed.
Please sign-up to scribe for a lecture using this sign-up spreadsheet.
Title | File | |
1 | Mistake Bound Analysis (2012/03/27) | |
2 | The Online Optimization Game (2012/03/29) | PDF (tex) |
3 | Follow-The-(Regularized)-Leader (2012/04/03) | PDF (tex) |
4 | Linearizing Convex Losses (2012/04/05) | PDF (tex) |
5 | Online Subgradient Descent (OGD) (2012/04/10) | PDF (tex) |
6 | Adaptive Learning Rates for OGD (2012/04/12) | PDF (tex) |
7 | Adaptive Per-Coordinate Learning (2012/04/17) | PDF (tex) |
8 | The FTRL Algorithm with Strongly Convex and L1 Regularizers (2012/04/19) | PDF (tex) |
9 | Guest Lecture (2012/04/24) | |
10 | More Algorithms using Strongly Convex Regularizers (2012/04/26) | |
11 | (2012/05/01) | |
12 | Partial Information: Bandit Problems and EXP3 (2012/05/03) | |
13 | Contextual Bandits / Bandits with Expert Advice (2012/05/08) | PDF (tex) |
14 | Analysis of EXP4 with Loses (2012/05/10) | PDF (tex) |
15 | ||
16 | Combinatorial Bandits | |
17 | ||
18 | ||
19 | ||
20 | References and Other Topics (2012/05/31) |