Machine Learning Foundations, Fall 2017

Course Description

Machine learning is the study that allows computers to adaptively improve their performance with experience accumulated from the data observed. The course teaches the most fundamental algorithmic, theoretical and practical tools that any user of machine learning needs to know.


Course Information


Class Policy

Course Plan (tentative)

datesyllabustodo/donesuggested reading
09/12course introduction;
topic 1: when can machines learn?
course slides
09/19the learning problemcourse slides; LFD 1.0, 1.1.1, 1.2.4
09/26learning to answer yes/no
types of learning
course slides; course slides; LFD 1.1.2, 3.1; LFD 1.2
10/03(unfinished parts last week); feasibility of learninghomework 1 announcedcourse slides; LFD 1.3
10/10no class because of double-ten holiday
10/17no class because of DSAA conference
10/24topic 2: why can machines learn?
(unfinished parts last week); training versus testing
course slides; LFD 2.0, 2.1.1
10/31(unfinished parts last week); theory of generalizationcourse slides; LFD 2.1.2
11/07no class because of midterm (good luck!)
11/14(unfinished parts last week); the VC dimensionhomework 1 due; homework 2 announcedcourse slides; LFD 2.2
11/21(unfinished parts last week); noise and errorcourse slides; LFD 1.4
11/28topic 3: how can machines learn?
linear regression
course slides; LFD 3.2
12/05logistic regression course slides; LFD 3.3
12/12linear models for classification homework 2 duecourse slides; LFD 3.3 (for SGD part only)
12/19nonlinear transformation homework 3 announcedcourse slides; LFD 3.4
12/26topic 4: how can machines learn better?
hazard of overfitting
course slides; LFD 4.0, 4.1
01/02regularization; validation homework 3 due;
homework 4 announced
course slides; course slides; LFD 4.2, 4.3
01/09three learning principle course slides; LFD 5
01/16winter vacation begins (really?)homework 4 due