Inhalt

•  
   Introduction
•  
   Background
•  
   Structure
•  
   Part I General Background
•  
  Background and Related Work
•  
   Multiagent Systems
•  
   The Online Partitioning Problem
•  
   Multirobot Task Allocation
•  
   Swarm Intelligence
•  
   Transfer Learning
•  
   Optimization
•  
  Reinforcement Learning
•  
   Introduction
•  
   Single Agent Reinforcement Learning
•  
   Multiagent Reinforcement Learning
•  
   Reward Shaping
•  
   Conclusion
•  
   Part II Foundations
•  
  Sequential Stage Games and Engineered Rewards
•  
   Sequential Stage Games
•  
   Engineered Reward-Based Models
•  
  The Iterative Agent Partitioning Problem
•  
   Problem Motivation
•  
   Formal Definition
•  
   Properties
•  
   Message Based Optimal Algorithms
•  
   Complexity
•  
   Summary
•  
  Storage Media
•  
   Concept
•  
   Strategic Positioning
•  
   Influence of Obstacles
•  
   Greedy Placement Strategy
•  
   Summary
•  
   Part III Algorithms
•  
  Reinforcement Learning with Noisy Rewards
•  
   Basic Definitions
•  
   Deterministic Q-Learning with Noisy Rewards
•  
   Q-Learning with Noisy Rewards and Optimistic Updates
•  
   Conclusion
•  
   Distributed Stateless Learning
•  
   Introduction
•  
   Approach
•  
   Formal Analysis
•  
   Evaluation
•  
   Complexity
•  
   Distributed Stateless Learning with Noisy Rewards
•  
   Discussion
•  
   Coordination in Sequential Stage Games
•  
   Coordination Strategies
•  
   Modeling the IAPP as Sequential Stage Game
•  
   Summary
•  
   Empirical Analysis
•  
   Settings
•  
   Basic Distributed Stateless Learning
•  
   Strategy Specific Simulations
•  
   Reward Types and Coordination Strategies
•  
   Evaluation in Different Scenarios
•  
   Discussion
•  
   Storage Medium-Based Distributed Stateless Learning
•  
   Approach
•  
   Empirical Analysis
•  
   Conclusion
•  
   Region-Based Heuristics
•  
   Target-Regions
•  
   Approximation of Target-Regions
•  
   Voronoi Regions
•  
   Discussion
•  
   Part IV Conclusion
•  
  Conclusions and Future Work
•  
   Conclusions
•  
   Future Work
•  
   Part V Appendix
•  
   Medium Coverage
•  
   Computation of Medium Coverage
•  
   Optimal Coverage
•  
   Simulation Results
•  
   Simulation Details for Global Reward Settings
•  
   Simulation Details for Local Reward Settings
•  
   Simulation Details for Locally Transformed Global Rewards
•  
   List of Notations
•  
   Abbreviations
•  
   References