Offline Risk-sensitive RL with Partial Observability to Enhance Performance in Human-Robot Teaming.
Giorgio AngelottiCaroline P. C. ChanelAdam H. M. PintoChristophe LounisCorentin ChauffautNicolas DrougardPublished in: CoRR (2024)
Keyphrases
- partial observability
- risk sensitive
- human robot
- reinforcement learning
- markov decision processes
- partially observable
- model free
- action selection
- optimal control
- markov decision problems
- markov decision process
- human robot interaction
- dialogue system
- control policies
- state space
- belief state
- learning agent
- planning problems
- reinforcement learning algorithms
- function approximation
- optimal policy
- infinite horizon
- utility function
- humanoid robot
- partially observable markov decision processes
- finite state
- decision theoretic
- temporal difference
- expected utility
- robotic systems
- decision problems
- average cost
- heuristic search
- policy iteration
- supervised learning
- dynamic programming
- multi agent
- markov chain
- action space
- decision making
- artificial intelligence
- learning process
- domain knowledge
- planning under uncertainty
- evolutionary computation
- human users