Fully asynchronous policy evaluation in distributed reinforcement learning over networks

Keyphrases

• Policy Evaluation 100%
• Distributed Reinforcement Learning 100%
• Networked Learning 100%
• Numerical Experiments 50%
• Value Function 50%
• Peer-to-peer Networks 50%
• Multiple Time Scales 50%
• Node number 50%
• Gradient-based 50%
• Push-pull 50%
• Distributed Algorithms 50%
• Decision Problems 50%
• Evaluation Challenges 50%
• Delayed Information 50%
• Asynchronous Methods 50%
• Linear Growth Rate 50%
• Gossip Based 50%
• Graphical Approach 50%
• Asynchronous Setting 50%
• Stragglers 50%
• Augmented Graph 50%
• Incremental Aggregated Gradient 50%
• Exact Convergence 50%

Computer Science

• Reinforcement Learning 100%
• Speed-up 50%
• Function Value 50%
• Decision Problem 50%
• Peer to Peer Networks 50%
• Distributed Algorithm 50%