Lecture Number | Date | Topic | Slides | Readings |
---|---|---|---|---|
1 | 9/7/2023 | Logistics and Intro, Perceptron | PDF, PDF | Lecture 1 of Yaoliang's notes UML Section 9.1 ESL Section 4.5 |
2 | 9/12/2023 | Finish Perceptron | Same as last lecture | |
3 | 9/14/2023 | Start Linear Regression | UML Section 9.2, 11.2 ESL Section 3.2, 3.4, 7.10 ISL Section 3.1-3.2, 5.1, 6.2 Calculus derivation |
|
4 | 9/19/2023 | Finish Linear Regression, k-Nearest Neighbour Classification | PDF, PDF | UML Section 19 ESL Section 2.3.2, 13.3 ISL Section 2.2 |
5 | 9/21/2023 | Logistic Regression | UML Section 9.3 ESL Section 4.4 ISL Section 4.3 |
|
6 | 9/26/2023 | Hard-Margin Support Vector Machines | UML Section 15 ESL Section 12.1-12.3 ISL Section 9.1-9.2 |
|
7 | 9/28/2023 | Soft-Margin Support Vector Machines, start Kernels | Last lecture's SVM content UML Section 16 ESL Section 12.3 ISL Section 9.3 |
|
8 | 10/03/2023 | Finish Kernels, start Decision trees | Last lecture's Kernels content UML Section 18 ESL Section 9.2 ISL Section 8.1 |
|
9 | 10/05/2023 | Finish Decision trees, Bagging | Last lecture's Decision Tree content UML Section 10 ESL Section 8.2, 8.7, 10 ISL Section 8.2 |
|
10 | 10/17/2023 | Boosting | Last lecture's Boosting content | |
11 | 10/19/2023 | Multilayer Perceptrons | DL Section 6 D2L Section 4 UML Section 20 ISL Section 10.1, 10.2, 10.7 3Blue1Brown videos on backprop: 1, 2 |
|
12 | 10/24/2023 | Deep Networks | DL Section 7, 8 D2L Section 5, 11 An overview of gradient descent optimization algorithms, by Sebastian Ruder |
|
13 | 10/26/2023 | Optimization, Convolutional Neural Networks | DL Section 9 D2L Section 6, 7 ISL Section 10.3 |
|
14 | 10/31/2023 | Finish Convolutional Neural Networks | DL Section 9 D2L Section 6, 7 ISL Section 10.3 |
|
15 | 11/02/2023 | Recurrent Neural Networks | DL Section 10 D2L Section 8, 9 ISL Section 10.5 |
|
16 | 11/07/2023 | k-Means and Gaussian Mixture Models |
UML Section 22.2, 24.4 ESL Section 6.8, 8.5, 14.3 ISL Section 10.3 |
|
17 | 11/09/2023 | Finish GMMs, Autoencoders |
DL Section 14, 20 Building Autoencoders in Keras, by Francois Chollet An Introduction to Variational Autoencoders, by Diederik Kingma and Max Welling |
|
18 | 11/14/2023 | Variational Autoencoders | Same as last time | |
19 | 11/16/2023 | Generative Adversarial Networks | Yaoliang used his slides, mine are PDF |
DL Section 20 D2L Section 17 |
20 | 11/21/2023 | Robustness | Yaoliang used his slides, mine are PDF | Adversarial Robustness - Theory and Practice by Zico Kolter and Aleksander Madry |
21 | 11/23/2023 | Flows | Video | |
22 | 11/28/2023 | Privacy | Notes | |
23 | 11/30/2023 | Attention | The Annotated Transformer, by Sasha Rush D2L Section 10 Attention Is All You Need Improving Language Understanding with Unsupervised Learning BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding Better Language Models and Their Implications Language Models are Few-Shot Learners |
|
24 | 12/05/2023 | Ethics | Private traits and attributes are predictable from digital records of human behavior Sharing learnings about our image cropping algorithm Gender Shades Machine Bias: There’s software used across the country to predict future criminals. And it’s biased against blacks. |