Towards Human-Like RL: Taming Non-Naturalistic Behavior in Deep RL via Adaptive Behavioral Costs in 3D Games.
Kuo-Hao HoPing-Chun HsiehChiu-Chou LinYou-Ren LuoFeng-Jian WangI-Chen WuPublished in: ACML (2023)
Keyphrases
- reinforcement learning
- human behavior
- learning agents
- adaptive behavior
- selective perception
- transfer learning
- reinforcement learning algorithms
- model free
- state space
- function approximation
- behavioral model
- markov decision processes
- action selection
- adaptive control
- learning algorithm
- agent behavior
- multi agent
- machine learning
- mental models
- total cost
- temporal difference
- video games
- nash equilibria
- game theory
- optimal policy
- reinforcement learning methods
- human decision making
- learning process
- reinforcement learning agents