This is an old revision of the document!
March 20: A4grader.tar linked to on the A4 web page has been updated. It longer checks for QDA-related functions.
March 18: There will be no lecture class on Wednesday, March 22nd. Chuck's office hours on March 22nd are cancelled.
Lecture videos are available at this CS480 video recordings site.
| Week 1:|
Jan 17 - Jan 20
|Overview. Intro to machine learning. Python.|| 01 Course Overview,|
02 Matrices and Plotting,
| The Great A.I. Awakening, by Gideon Lewis-Krause, NYT, Dec 14, 2016.|
Section 1 of Scipy Lecture Notes
| Week 2:|
Jan 23 - Jan 27
|Probability distributions and regression.|| 03 Linear Regression,|
04 Gaussian Distributions
| Week 3:|
Jan 30 - Feb 3
|Probabilistic Linear Regression. Ridge regression. Data partitioning. On-line, incremental regression.|| 05 Fitting Gaussians,|
06 Probabilistic Linear Regression,
07 Linear Ridge Regression and Data Partitioning,
08 Sample-by-Sample Linear Regression
|A1 Linear Regression due Monday, January 30th at 10:00 PM.|
| Week 4:|
Feb 6 - Feb 10
| Regression with fixed nonlinearities. Nonlinear regression with neural networks.|
Feb 10: Guest Speaker Mike Morain, Machine Learning at Amazon, UK.
| 09 Linear Regression with Fixed Nonlinear Features,|
10 Nonlinear Regression with Neural Networks
| Week 5:|
Feb 13 - Feb 17
|Neural Networks|| 10 Nonlinear Regression with Neural Networks,|
11 More Nonlinear Regression with Neural Networks
| A2 Ridge Regression with K-Fold Cross-Validation due Monday, February 13th at 10:00 PM.
Here are examples of good A2 reports.
| Week 6:|
Feb 20 - Feb 24
|Neural Networks. Autoencoders. Guest lectures by our GTA, Jake Lee.||12 Autoencoder Neural Networks|
| Week 7:|
Feb 27 - Mar 3
| Recurrent Neural Networks.|
Conditional probabilities and Bayes Rule
| 13 Recurrent Neural Networks|
14 Introduction to Classification
| A3 Neural Network Regression due Wednesday, March 1st at 10:00 PM.
Here are examples of good A3 reports.
| Week 8:|
Mar 6 - Mar 10
|Classification. LDA and QDA. Linear and Nonlinear Logistic Regression.|| 15 Classification with Linear Logistic Regression|
16 Classification with Nonlinear Logistic Regression Using Neural Networks
| Week 9:|
Mar 20, Mar 24
No class March 22nd.
|Classification. Analysis of Trained Networks. Bottleneck Networks. Hand-Drawn Digit Classification.|| 17 Analysis of Neural Network Classifiers and Bottleneck Networks|
| Week 10:|
Mar 27 - Mar 31
|Convolutional Neural Networks. Reinforcement Learning.|| 19 Convolutional Neural Networks|
20 Introduction to Reinforcement Learning
|Reinforcement Learning: An Introduction, by Richard Sutton and Andrew Barto. 2nd edition draft. On-line and free.|
| Week 11:|
Apr 3 - Apr 7
|Reinforcement Learning. Two-player games.||21 Reinforcement Learning for Two Player Games||Reinforcement Learning: An Introduction, by Richard Sutton and Andrew Barto. 2nd edition draft. On-line and free.|| A4 Classification with LDA and Logistic Regression due Wednesday, April 5th at 10:00 PM.
Here are examples of good A4 reports.
Project Proposal due Friday, April 7th at 10:00 PM.
| Week 12:|
Apr 10 - Apr 14
|Neural networks as Q functions.|| 22 Reinforcement Learning with Neural Network as Q Function|
Faster RL by Pre-training
| The Dark Secret at the Heart of AI|
The Tiny Changes That Can Cause AI to Fail
| Week 13:|
Apr 17 - Apr 21
|Unsupervised Learning. Dimensionality reduction.|| 23 Linear Dimensionality Reduction|
24 Nonlinear Dimensionality Reduction with Digits Example
25 K-Means Clustering
26 Hierarchical Clustering
| Week 14:|
Apr 24 - Apr 28
|A5 Control a Marble with Reinforcement Learning due Monday, April 24th at 10:00 PM.|
| Week 15:|
May 1 - May 5
| Finals Week:|
May 8 - May 11
|Final project due Tuesday, May 9, 10:00 PM. Details on report requirements will be posted here soon.|