Course Description
Advanced representation learning algorithms in machine learning, from the traditional subspace learning models to the recent deep representation learning models. Applications in the fields of computer vision, data mining, and natural language processing will be covered.
Athena Title
Advanced Representation Learn
Prerequisite
(CSCI 4380/6380 and CSCI(PHIL) 4550/6550) or permission of department
Semester Course Offered
Not offered on a regular basis.
Grading System
A - F (Traditional)
Course Objectives
The purpose of this course is to familiarize students with advanced topics in representation learning, from the traditional subspace learning models to the recent deep representation learning models. The course will also cover a brief review of linear algebra, convex optimization, and machine learning. Real-world applications in the fields of computer vision, data mining, and natural language processing will be introduced as well. The course is appropriate for graduate students who want to do research in machine learning, or explore representation learning algorithms for their research.
Topical Outline
1. Review of linear algebra and convex optimization 2. Machine learning overview 3. Subspace learning 3.1 Linear subspace learning 3.2 Nonlinear subspace learning 3.3 Multi-view subspace learning 3.4 Transfer subspace learning 3.5 Applications in computer vision 4. Deep representation learning 4.1 Word embedding, sentence embedding, and document embedding 4.2 Adversarial training and virtual adversarial training 4.3 Deep clustering 4.4 Interpretable representation learning 4.5 Applications in natural language processing and recommender system 5. Graph Learning 5.1 Graph construction 5.2 Graph convolutional networks 5.3 Applications in data mining and computer vision
Syllabus