STARC: A General Framework For Quantifying Differences Between Reward Functions.
Joar Max Viktor SkalseLucy FarnikSumeet Ramesh MotwaniErik JennerAdam GleaveAlessandro AbatePublished in: ICLR (2024)
Keyphrases
- reward function
- inverse reinforcement learning
- statistically significant
- reinforcement learning
- state space
- optimal policy
- markov decision processes
- state variables
- transition probabilities
- multiple agents
- reinforcement learning algorithms
- simple examples
- transition model
- policy search
- image segmentation
- active learning