Week | Day | Date | Subject (and lecture notes) | Readings |
1 | W | 1/24 | course overview |
| F | 1/26 | ML paradigms, feature-space representations | Mitchell chapter 1 or Murphy chapter 1 or James et al. chapter 1 and section 2.1; Dietterich, Nature Encyclopedia of Cognitive Science, 2003 |
2 | M | 1/29 | review of probability
decision tree learning, overfitting | Mitchell chapter 3, Murphy 16.2, James et al. chapter 8 |
| W | 1/31 | decision tree learning | Page and Ray, IJCAI 2003 |
| F | 2/2 | instance-based learning, k-nearest neighbor | Mitchell chapter 8 |
3 | M | 2/5 | machine learning methodology (part 1) | Mitchell chapter 5, Murphy 5.7.2, 6.5.3 Manning et
al., Sections 8.3-8.4 |
| W | 2/7 | machine learning methodology (part 2) | Provost et al. ICML 1998 |
| F | 2/9 | linear and logistic regression | James et al. 3.1-3.5, 6.2, Shalev-Shwartz and Ben-David 9.2-9.3 |
4 | M | 2/12 | Bayesian network learning (part 1) | Mitchell chapter 6, Murphy chapter 10 |
| W | 2/14 | Bayesian network learning (part 2) |
5 | M | 2/19 | Bayesian network learning (part 3) (updated 3/7) | Friedman et al. Machine Learning 1997 Friedman et al. UAI 1999 |
| W | 2/21 | neural network foundations | Mitchell chapter 4, Murphy 16.5, 28 LeCun et al., Nature, 2015 |
6 | M | 2/26 | neural network foundations | |
| W | 2/28 | deep neural networks (part 1) | Deep Learning tutorials by R. Salakhutdinov (scroll down to schedule for 1/26/17) |
| W | 2/28 | deep neural networks (part 2) click here for Animations from this lecture | |
7 | M | 3/5 | deep neural networks (part 3) (updated 3/28) | |
8 | M | 3/12 | learning theory: PAC model | Mitchell chapter 7 |
| W | 3/14 | learning theory: mistake-bound model learning theory: bias-variance decomposition | Geman et al., Neural Computation, 1992 (Sections 1-3) |
9 | M | 3/19 | support vector machines | Ben-Hur and Weston, 2010 |
| W | 3/21 | support vector machines |
10 | M | 4/2 | ensemble methods | Dietterich, AI Magazine 1997 (through page 105) |
| W | 4/4 | reinforcement learning | Mitchell chapter 13 |
11 | M | 4/9 | Learning to play world-class Go, Chess, and Shogi in 24 hours via reinforcement learning and DNNs |
AlphaZero, which is based on AlphaGo Zero, which is based on AlphaGo |
12 | M | 4/16 | Undirected Probabilistic Graphical Models | |
| W | 4/18 | rule learning and relational learning | Quinlan, Machine Learning 1990, Domingos and Richardson, In ISRL, 2007 |
13 | M | 4/30 | privacy and fairness |