Safe Reinforcement Learning for Arm Manipulation with Constrained Markov Decision Process.
Patrick AdjeiNorman L. TasfiSantiago Gomez-RoseroMiriam A. M. CapretzPublished in: Robotics (2024)
Keyphrases
- markov decision process
- reinforcement learning
- state space
- optimal policy
- markov decision processes
- temporal difference learning
- finite horizon
- policy iteration
- action space
- initial state
- infinite horizon
- markov games
- state action
- reward function
- dynamic programming
- function approximation
- transition probabilities
- learning algorithm
- hierarchical reinforcement learning
- control problems
- planning problems
- stationary policies
- partial observability
- partially observable
- machine learning
- markov chain
- probability distribution
- optimal control
- search space
- state variables
- temporal difference