CS 158: Machine Learning
Harvey Mudd College

This material is from the Fall 2017 offering of the course. The syllabus, lecture notes, and homeworks that are provided to students are available via links. Additional resources available upon request.

Wk Date Topic Stanford ML Notes [recommended] Readings [recommended] Assignments Project
1 W 08/30 - Introduction - Lec 1 (Pgs 1-2) - The Discipline of Machine Learning; Daume 1-1.2; LfD 1-1.2 PS 1 out [Setup] [Written] [Programming]  
2 M 09/04 - Decision Trees   - Daume 1.3-1.10; Flach 5-5.1    
W 09/06 - k-Nearest Neighbors
- Evaluation: Metrics
  - Daume 2-2.3, 2.5-2.6; Flach 8-8.3
- Daume 4.5; LfD 4.3; Flach 2.1,12
PS 1 due
PS 2 out [Written] [Programming]
 
3 M 09/11 - Evaluation: Protocol
- Linear Regression Setup
- None
- Lec 5 (Pgs 2-4) + Lec 1 (Pgs 3-13)
- Daume 4.6-4.7
- Daume 6-6.2, 6.4-6.6; LfD 3.2-3.2.1; Flach 7.1
   
W 09/13 - Linear Regression
- Regularization
- (see above)
- None
- (see above)
- Daume 6.3; LfD 4-4.2
PS 2 due
PS 3 out [Written] [Programming]
 
4 M 09/18 - Logistic Regression - Lec 1 (Pgs 16-19) - LfD 3.3    
W 09/20 - Perceptron - Lec 1 (Pg 19) + Lec 6 - Daume 3; Flach 7.2 PS 3 due
PS 4 out [Written] [Programming]
 
5 M 09/25 - Support Vector Machines - Lec 3 (Pgs 1-7, 19-20; optional 7-13) - Daume 6.7; LfD 3.4; Flach 7.3    
W 09/27 - Kernels - Lec 3 (Pgs 13-29) - Daume 9-9.2, 9.4-9.6; Flach PS 4 due
PS 5+6 out [Written] [Programming]
   
6 M 10/02 - Advanced Evaluation Metrics
- Imbalanced Data
  - Daume 4.5 (review)
- Daume 5-5.1; Flach 12.1
   
W 10/04 - Multiclass Classification
- Advice for Applying ML
- None
- "Advice for Applying ML"
- Daume 5.2; Flach 3.1
- None
PS 5 due
PS 6 (Parts 1-3) "due"
 
7 M 10/09 - Ensemble Methods: Bagging   - Daume 11-11.1, 11.3; Flach 11-11.1   Project out
W 10/11 Midterm (in-class)       [brainstorm projects]
8 M 10/16 Fall Break
W 10/18 - Ensemble Methods: Boosting   - Daume 11.2; Flach 11.2 PS 6 due
PS 7 out [Written] [Programming]
[brainstorm projects]
9 M 10/23 - Dimensionality Reduction (PCA) - Lec 10 - Daume 13.2; Flach 10.3    
W 10/25 - Clustering - Lec 7a + Lec 7b - Daume 2.4, 13-13.1; Flach 3.3, 8.4-8.5 PS 7 due
PS 8 out [Programming]
Proposal Conference (by 5pm)
10 M 10/30 Project Proposal Presentations        
W 11/01 Project Proposal Presentations       Proposal Writeup due Fri (5pm)
11 M 11/06 - Gaussian Mixture Models
- Expectation Maximization
- Sec Notes 7 (Pgs 1-2)
- Lec 8 (Pgs 1-6)
- Daume 14-14.1; Flach 9.4
- Daume 14.2
   
W 11/08 - EM Applied to GMMs
- Hidden Markov Models Overview
- Lec 8 (Pgs 6-8)
- Sec Notes 6 (Pgs 1-6)
- None
- Rabiner Tutorial (Pgs 257-262)
PS 8 due
PS 9 out [Written]
 
12 M 11/13 - HMMs: Inference - Sec Notes 6 (Pgs 6-8) - Rabiner Tutorial (Pgs 262-264)    
W 11/15 - HMMs: Learning - Sec Notes 6 (Pgs 8-13) - Rabiner Tutorial (Pgs 264-266)   Status Update due
13 M 11/20 Project Conferences        
W 11/22 - Computational Learning Theory: Finite Hypothesis Spaces - Lec 4 (Pgs 1-8) - Daume 10.1-10.5; LfD 1.3, 2.2-2.3 PS 9 due
PS 10 out [Written]
 
14 M 11/27 - Computational Learning Theory: Infinite Hypothesis Spaces - Lec 4 (Pgs 8-11) - Daume 10.6; LfD 2.1    
W 11/29 - Special Topics
- Course Evaluations
       
15 M 12/04 Project Presentations        
W 12/06 Project Presentations     PS 10 due Project Work Log due