GENESIS-RL: GEnerating Natural Edge-cases with Systematic Integration of Safety considerations and Reinforcement Learning.
Hsin-Jung YangJoe BeckMd. Zahid HasanEkin BeyazitSubhadeep ChakrabortyTichakorn WongpiromsarnSoumik SarkarPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- learning algorithm
- model free
- state space
- action space
- markov decision processes
- temporal difference
- autonomous learning
- dynamic programming
- multi agent
- machine learning
- control problems
- reinforcement learning methods
- rl algorithms
- transfer learning
- edge information
- partially observable domains
- data integration
- state action
- edge detection
- continuous state and action spaces
- direct policy search
- markov decision problems
- partially observable
- multiscale
- learning problems