Finite-Time Analysis of Decentralized Temporal-Difference Learning with Linear Function Approximation

Keyphrases

• Decentralized Setting 33%
• Engineering Application 33%
• Error Bound 33%
• Evaluation Challenges 33%
• Finite-sample Analysis 33%
• Finite-time Analysis 100%
• Fully Decentralized 66%
• Large State Space 33%
• Learning Algorithm 33%
• Linear Function Approximation 100%
• Local Estimates 66%
• Lyapunov Analysis 33%
• Markovian 66%
• Multi-agent Reinforcement Learning 33%
• Policy Evaluation 33%
• Projection Step 33%
• Shared Environment 33%
• Temporal Difference 33%
• Temporal Difference Learning 100%
• Theoretical Understanding 33%
• Value Function 33%

Computer Science

• Engineering Application 33%
• Function Approximation 100%
• Large State Space 33%
• Learning Algorithm 33%
• Linear Function 100%
• Multi-Agent Reinforcement Learning 33%
• Shared Environment 33%
• temporal difference 33%
• temporal difference learning 100%

Mathematics

• Data Sample 50%
• Error Bound 50%
• Finite Time 100%
• Linear Function 100%