Intelligent agents

The course will contain the following chapters:

1. Introduction and overview
   • definition of an agent
   • intelligent agents
   • rational agents
2. Reactive agents
   • Reactive agents
   • Subsumption architecture
   • Decision processes
   • Markov decision processes (MDP)
3. Learning Agents/reinforcement learning

• Learning without state; exploration/exploitation tradeoff
• Adversarial learning
• Learning with state

Deliberative agents/game-playing

• Deliberative Agents
• State-based planning
• Planning by search
• Planning with an adversary

Planning with factored representations

• World models with factored representation
• Non-linear planning
• Planning graphs
• Planning as SAT/CSP
• Planning and MDP

Deep learning agents

• Model-free reinforcement learning
• Off-policy learning
• Multiagent reinforcement learning

Multi-agent Systems

• Introduction
• Ontologies
• Agent communication languages
• Interaction protocols

Distributed coordination

• Hierarchical coordination
• Peer-to-peer coordination
• Social laws
• Cooperation protocols
• Explicit cooperative planning
• Distributed constraint satisfaction
  

Self-interested agents (Game Theory)

• Games
• Pure and mixed strategies
• Equilibrium, Nash equilibrium
• N-Players games
• Social choice problems

Agent negotiation

• Negotiation protocols
• Conflict Situations
• Negotiation as optimization
• VCG tax

Auctions and Mechanism Design

• Auction protocols and bidding strategies
• Truthful mechanisms; VCG mechanisms
• Truthful information elicitation

Coalitions and Group Decisions

• Cooperative Game Theory
• Shapley Values
• Social Choice Problems
  
• Voting and Manipulation

Agent platforms and environments, Applications of agents

• An introduction to the Java Agent Development Framework

• Implementing agents with web services

Final Exam