Deep Learning and Unsupervised Feature Learning
 Winter 2011


Project Description

In this class, we will develop unsupervised deep learning algorithms that are capable of learning useful features for a range of machine learning applications. Unlike most previous CS294A's, this course will pursue work in developing new machine learning algorithms (i.e., "core" or "algorithmic" machine learning) rather than in "applied" machine learning.

Because it is challenging to work on algorithmic machine learning, we will be able to work with only a small number of students, and enrollment will be limited. You can find details about enrollment requirements on the Handouts and enrollment information page.

Course Information

Course Instructor:

Andrew Ng. (

Class meetings:

This is a project course. There will be no weekly lectures, and only two introductory homeworks. We will spend the quarter working in teams on different deep learning related projects.

The whole class will meet on 5th January (4.15pm-5.30pm, Gates 120) and 9th Feb (4.15-5.30pm).

Final project presentations will be held Thursday, March 17, 1:30pm-3:00pm.

If you were unable to attend the first meeting but would like to take CS294A, please email to let us know.

More information: