A Novel Ping-pong Task Strategy Based on Model-free Multi-dimensional Q-function Deep Reinforcement Learning.
Hongxu MaJianyin FanQiang WangPublished in: ICSAI (2022)
Keyphrases
- model free
- reinforcement learning
- multi dimensional
- ping pong
- reinforcement learning algorithms
- function approximation
- function approximators
- temporal difference
- control policy
- policy iteration
- learning algorithm
- rl algorithms
- state space
- average reward
- machine learning
- transfer learning
- policy evaluation
- impedance control
- evaluation function
- learning problems
- markov decision processes
- supervised learning
- least squares
- reinforcement learning methods
- dynamic programming