Multi-Stage Reinforcement Learning for Non-Prehensile Manipulation.
Dexin WangFaliang ChangChunsheng LiuPublished in: CoRR (2023)
Keyphrases
- multistage
- reinforcement learning
- optimal policy
- dynamic programming
- production system
- state space
- single stage
- lot sizing
- stochastic optimization
- markov decision processes
- stochastic programming
- function approximation
- model free
- temporal difference
- markov decision process
- finite horizon
- reinforcement learning algorithms
- optimal control
- production line
- learning algorithm
- probability distribution
- finite state
- long run
- reward function