MAP Inference for Bayesian Inverse Reinforcement Learning.
Jaedeug ChoiKee-Eung KimPublished in: NIPS (2011)
Keyphrases
- map inference
- inverse reinforcement learning
- graphical models
- markov random field
- image labeling
- bayesian networks
- preference elicitation
- reward function
- semantic segmentation
- probabilistic model
- maximum likelihood
- random variables
- belief propagation
- posterior distribution
- decision theory
- posterior probability
- cutting plane
- approximate inference
- maximum a posteriori
- higher order
- temporal difference
- dynamic systems
- parameter estimation
- graph cuts
- computational complexity