Closing the Dynamics Gap via Adversarial and Reinforcement Learning for High-Speed Racing.
Jingyu NiuYu HuWei LiGuangyan HuangYinhe HanXiaowei LiPublished in: IJCNN (2022)
Keyphrases
- high speed
- reinforcement learning
- low power
- multi agent
- markov decision processes
- high speed networks
- dynamic model
- function approximation
- state space
- dynamical systems
- policy search
- real time
- temporal difference
- robot control
- action selection
- reinforcement learning algorithms
- optimal policy
- model free
- game theory
- function approximators
- multi agent reinforcement learning
- robotic control
- morphological operators
- frame rate
- transfer learning
- monte carlo
- supervised learning
- hidden markov models
- decision trees
- learning algorithm
- machine learning
- neural network