Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for Robotics Control with Action Constraints.
Kazumi KasauraShuwa MiuraTadashi KozunoRyo YonetaniKenta HoshinoYohei HosoePublished in: CoRR (2023)
Keyphrases
- reinforcement learning algorithms
- actor critic
- reinforcement learning
- optimal control
- temporal difference
- policy gradient
- markov decision processes
- state space
- model free
- action selection
- reinforcement learning problems
- control problems
- function approximation
- learning algorithm
- reinforcement learning methods
- temporal difference learning
- stochastic games
- reward function
- initial state
- control system
- approximate dynamic programming
- control strategies
- adaptive control
- neuro fuzzy
- action space
- dynamic programming
- control strategy
- learning tasks
- learning agent
- dynamic environments
- evaluation function
- planning problems
- state action
- infinite horizon