B553: Spring 2009
Neural and Genetic Approaches to Artificial Intelligence
Indiana University Computer Science Department

Where, when, who, how

Section: 26212

Time and place

TuTh, 9:30 - 10:45, I107

Instructor: Mike Gasser

• Office hours: M 10:00-12:30, Eigenmann 804; Th 11:00-12:30, Lindley 230H
• Email (please put "B553" in the subject line)

Associate Instructor: Giancarlo Schrementi

• Office hours: TBA
• Email: gischrem AT indiana DOT edu

Grades

Oncourse site

Overview and prerequisites

This course covers the most important computational mechanisms for implementing intelligent behavior that are based on biological systems, including individual animals and populations of animals. Topics will include:

• Simple stimulus-response agents
• Evolutionary computation
• Reinforcement learning
• Content-addressable memories
• Competitive learning
• Dimensionality reduction
• Self-organizing maps
• Modular and constructivist architectures
• Sequence learning

The overall approach will be to start with simple agents in a simple artificial two-dimensional world and to make them smarter by causing them to learn, to evolve, and to re-represent their inputs in efficient ways.

Students are expected to have taken a graduate or advanced undergraduate artificial intelligence course.

Readings

All readings for the course will be available online.

© 2009. Indiana University and Michael Gasser. www.cs.indiana.edu/classes/b553/index.html 2009-01-15