Toward Computationally Efficient Inverse Reinforcement Learning via Reward Shaping.
Lauren H. CookeHarvey KlyneEdwin ZhangCassidy LaidlawMilind TambeFinale Doshi-VelezPublished in: Tiny Papers @ ICLR (2024)
Keyphrases
- inverse reinforcement learning
- reward shaping
- reward function
- reinforcement learning
- reinforcement learning algorithms
- state space
- markov decision processes
- multiple agents
- partially observable
- optimal policy
- preference elicitation
- temporal difference
- complex domains
- markov decision problems
- markov decision process
- transition probabilities
- state variables
- multi agent
- markov chain
- learning agent
- probability distribution
- dynamic programming