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.
|09/13||no class because of mid-autumn festival|
topic 1: when can machines learn?
the learning problem
course slides; LFD 1.0, 1.1.1, 1.2.4
|09/27||learning to answer yes/no||course slides; LFD 1.1.2, 3.1|
|10/04||types of learning||homework 1 announced||course slides; LFD 1.2; LFD 1.3|
|10/05||no class on this make-up day because it will not be effective to take four hours within two days|
|10/11||no class because of long weekend of double-ten holiday|
feasibility of learning
||course slides; LFD 1.3|
topic 2: why can machines learn?
training versus testing;
theory of generalization
course slides; LFD 2.0, 2.1.1;
course slides; LFD 2.1.2
|11/01||the VC dimension||homework 1 due||course slides; LFD 2.2|
|11/08||noise and error||homework 2 announced||course slides; LFD 1.4|
|11/15||no class because of NTU birthday|
topic 3: how can machines learn?
|course slides; LFD 3.2|
|11/29||logistic regression||course slides; LFD 3.3|
|12/06||linear models for classification||course slides; LFD 3.3 (for SGD part only)|
|12/13||nonlinear transformation||homework 2 due||course slides; LFD 3.4|
topic 4: how can machines learn better?
hazard of overfitting
|homework 3 announced||course slides; LFD 4.0, 4.1|
|12/27||regularization||course slides; LFD 4.2|
|01/03||validation||course slides; LFD 4.3|
|01/10||three learning principle||course slides; LFD 5|
|01/14 (note: Tuesday)||winter vacation begins (really?)||homework 3 due|