“Equations are just the boring part of mathematics. I attempt to see things in terms of geometry.”
 Stephen Hawking (19422021)
Coordinates  [Syllabus] [Canvas] [Registrar] [Slack Channel] 
Instructor  Liang Huang (liang.huang@oregonstate.edu) 
TA & Office Hours 
Junkun Chen (chenjun2@oregonstate.edu); office hours: W/F 56pm on his zoom. Ning Dai (dain@oregonstate.edu); office hours: M 56pm on the above zoom. Instructor office hours available the weeks before HWs are due (usually 3pm Tuesdays); on the same zoom. 
Prerequisites 

Textbooks 

Grading 

Communication 
We use three tools for communication:

Week  Topics (CIML References)  Slides/Handouts  Videos  homework/exercises/quizzes 

Unit 1 (weeks 13): ML intro, kNN, and math/numpy review  
1  12. Introduction (§0, §2.7)
3. Training, Test, and Generalization Errors (§2.52.6), Underfitting and overfitting (§2.4), and Leaveoneout crossvalidation (§5.6). 4. knearest neighbor classifier (kNN) (§3) 5. viewing HW1 data on terminal 67. data preprocessing: binarization  slides (topics 15)
notebook (python3) (topics 67) and toy.txt 
background survey (required) Quiz 1 (ML basics) HW1 out (kNN) [tex] [data] [validate.py] [random_output.py]  
2 
Geometric Review of Linear Algebra Numpy Tutorial (also matplotlib): 1. ipython notebook; ndarray; %pylab; +//*, dot, concat 2. linear regression; np.polyfit; np.random.rand(); broadcasting 3. visualizing vectors operations; dot product, projection  notebook (python3) (topics 13) 
Quiz 2 (numpy/linear algebra)
you should finish at least parts 12 of HW1 by week 2.  
3  finish HW1!  HW1 DUE  
Unit 2 (weeks 45): Linear Classification and Perceptron Algorithm  
4  Linear Classification and Perceptron
1. Historical Overview; Bioinspired Learning (§4.1) 2. Linear Classifier (§4.34.4); Augmented space (not in CIML) 3. Perceptron Algorithm (§4.2) 4. Convergence Proof (§4.5) 5. Limitations and NonLinear Feature Map (§4.7, §5.4) 6. ipynb demo  slides
notebook (python3) (from [181])  HW2 out [tex] (same data as HW1)  
5  Perceptron Extensions; Perceptron in Practice 1. Python demo 2. Perceptron Extensions: voted and averaged (§4.6) 3. MIRA and aggressive MIRA (not in CIML) 4. Practical Issues (§5.15.4) 5. Perceptron vs. Logistic Regression (§9.6)  slides demo  HW2 due  
Unit 3 (weeks 67): Linear and Polynonmial Regression  
67  linear and polynomial regression (mostly not in CIML, but mentiond in §7.6)  HW3 (housing price prediction) kaggle data tex  
Unit 4 (weeks 89): Applications: Text Classification  
89  Application: Text Classification Sentiment Analysis (thumbs up?)  HW4 [tex] data and code  
Unit 5 (week 10): Exposure to cuttingedge ML research  
10  Paper Review: cuttingedge ML topics and papers  HW5 (paper review) 