Pruning With Scaled Policy Constraints for Light-Weight Reinforcement Learning.
Seongmin ParkHyungmin KimHyunhak KimJungwook ChoiPublished in: IEEE Access (2024)
Keyphrases
- lightweight
- reinforcement learning
- optimal policy
- policy search
- markov decision processes
- control policy
- reinforcement learning algorithms
- action selection
- partially observable environments
- communication infrastructure
- actor critic
- markov decision process
- policy iteration
- action space
- wireless sensor networks
- machine learning
- function approximation
- state and action spaces
- control policies
- partially observable
- state space
- reward function
- temporal difference
- markov decision problems
- pruning method
- search space