ME 538
AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS

Course Description

This course provides an introduction to autonomous agents and multiagent systems. In particular, it focuses on how to use agents as building blocks for different autonomous systems. Covered topics include reinforcement learning, game theory, swarms, auctions, and collectives. Because this course covers a constantly evolving field, there will be a significant paper reading component in addition to the regular lectures. Students are expected to spend at least three hours a week reading, discussing and critiquing assigned papers.

Topics

• Multiagent learning
• Agent objective functions
• Game theory
• Multiagent coordination

Learning Outcomes

The student, upon completion of this course, will be able to:

• Assess and describe the system characteristics needed for a distributed autonomous system to reach system level goals.
• Produce a reinforcement learning agent capable of goal driven behavior.
• Select the proper multiagent framework for a given domain.
• Derive agent utility/objective functions given a system level utility/objective function.
• Read and assess research papers in autonomous agents and multiagent systems.