Finite-Sample Analysis for Decentralized Batch Multiagent Reinforcement Learning with Networked Agents

Keyphrases

• Networked Agents 100%
• Multi-agent Reinforcement Learning 100%
• Finite-sample Analysis 100%
• Learning Environment 50%
• Finite Sample 33%
• Multi-agent Reinforcement Learning Algorithm 33%
• Error Analysis 16%
• Challenging Problems 16%
• Sampling Error 16%
• Competitive Agents 16%
• Number of Iterations 16%
• Error Term 16%
• Class Function 16%
• Zero-sum Game 16%
• Human-agent Teams 16%
• Statistical Accuracy 16%
• Reinforcement Learning 16%
• Time-varying Networks 16%
• Multiple Communities 16%
• Additional Error 16%
• Finite Sample Bounds 16%
• Theoretical Understanding 16%
• Community Understanding 16%
• Game Settings 16%
• Decentralized Computation 16%
• Cooperative Agents 16%
• Distributed multi-agent Reinforcement Learning 16%
• Action Value Function 16%

Computer Science

• Reinforcement Learning 100%
• multi-agent 100%

Mathematics

• Error Analysis 100%
• Error Term 100%