AD4RL: Autonomous Driving Benchmarks for Offline Reinforcement Learning with Value-based Dataset.
Dongsu LeeChanin EomMinhae KwonPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- autonomous driving
- grand challenge
- function approximation
- model free
- state space
- reinforcement learning algorithms
- markov decision processes
- stereo vision
- partially observable domains
- reinforcement learning methods
- temporal difference
- machine learning
- autonomous learning
- vision algorithms
- learning algorithm
- optimal policy
- urban traffic
- multi agent
- real time
- action selection
- action space
- optimal control
- rl algorithms
- markov decision process
- learning classifier systems
- actor critic
- state and action spaces
- direct policy search