Machine Learning, Fall 2021

Course Description

Machine learning allows computational systems to adaptively improve their performance with experience accumulated from the data observed. This course introduces the basics of learning theories, the design and analysis of learning algorithms, and some applications of machine learning.


Course Information


Class Policy

Course Plan (tentative)

09/23 course introduction;
topic 1: when can machines learn?
basics of machine learning
09/30 the learning problems
10/07 topic 2: why can machines learn?
feasibility of learning
homework 1 announced
10/14 theory of generalization
10/21 topic 3: how can machines learn?
linear models
homework 1 due; homework 2 released
10/28 beyond basic linear models
11/04 topic 4: how can machines learn better?
combatting overfitting
homework 3 released
11/11 combatting overfitting (2) homework 2 due
11/18 machine learning in practice final project released
11/25 topic 5: how can machines learn by embedding numerous features?
primal and dual support vector machine
homework 3 due; homework 4 released
12/02 soft-margin kernel support vector machine
12/09 no class because of NeurIPS 2021 homework 4 due; homework 5 released
12/16 topic 6: how can machines learn by combining predictive features?
decision tree, bagging, and random forest
12/23 adaptive boosting and gradient boosting homework 5 due; homework 6 released
12/30 topic 7: how can machines learn by distilling hidden features?
deep learning fundamentals
01/06 deep learning optimization homework 6 due
01/13 deep learning regularization and finale
01/20 no class and good luck with your final project final project due