Content Caching Policy Based on GAN and Distributional Reinforcement Learning.
Haipeng WengLixin LiQianqian ChengWei ChenZhu HanPublished in: ICC (2020)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- dynamic content
- action selection
- markov decision process
- co occurrence
- markov decision processes
- function approximation
- markov decision problems
- action space
- data access
- web content
- cache management
- actor critic
- approximate dynamic programming
- policy evaluation
- policy gradient
- reinforcement learning problems
- partially observable environments
- partially observable domains
- partially observable
- function approximators
- reward function
- multimedia content
- multi agent
- multimedia
- control policy
- state dependent
- metadata
- infinite horizon
- access patterns
- learning process
- website