2012
NeurIPS
NeurIPS 2012
Nonparametric Bayesian Inverse Reinforcement Learning for Multiple Reward Functions
Abstract
We present a nonparametric Bayesian approach to inverse reinforcement learning (IRL) for multiple reward functions. Most previous IRL algorithms assume that the behaviour data is obtained from an agent who is optimizing a single reward function, but this assumption is hard to be met in practice. Our approach is based on integrating the Dirichlet process mixture model into Bayesian IRL. We provide an efficient Metropolis-Hastings sampling algorithm utilizing the gradient of the posterior to estimate the underlying reward functions, and demonstrate that our approach outperforms the previous ones via experiments on a number of problem domains.
🌉
Interdisciplinary Bridge
- Artificial Intelligence and Reinforcement Learning
🐣
Hot Topic Early Bird
- reward function
Authors
Topics
Artificial Intelligence > Bayesian & Probabilistic > Bayesian Learning
Reinforcement Learning > Methods > Policy Learning
Machine Learning > Bayesian & Probabilistic > Bayesian Learning
Machine Learning > Bayesian & Probabilistic > Probabilistic Modeling
Machine Learning > Learning Types > Reinforcement Learning
Artificial Intelligence > Bayesian & Probabilistic > Bayesian Inference
Artificial Intelligence > Core AI > Reinforcement Learning
Machine Learning > Bayesian & Probabilistic > Markov Chain Monte Carlo
Keywords
dirichlet process
nonparametric bayesian
bayesian learning
bayesian inference
metropolis-hastings sampling
inverse reinforcement learning
markov chain monte carlo
reward function
bayesian inverse reinforcement learning
reward function estimation
multiple reward functions
dirichlet process mixture model