Timezone: »
Generative adversarial training for imitation learning has shown promising results on high-dimensional and continuous control tasks. This paradigm is based on reducing the imitation learning problem to the density matching problem, where the agent iteratively refines the policy to match the empirical state-action visitation frequency of the expert demonstration. Although this approach has shown to robustly learn to imitate even with scarce demonstration, one must still address the inherent challenge that collecting trajectory samples in each iteration is a costly operation. To address this issue, we first propose a Bayesian formulation of generative adversarial imitation learning (GAIL), where the imitation policy and the cost function are represented as stochastic neural networks. Then, we show that we can significantly enhance the sample efficiency of GAIL leveraging the predictive density of the cost, on an extensive set of imitation learning tasks with high-dimensional states and actions.
Author Information
Wonseok Jeon (KAIST)
Seokin Seo (KAIST)
Kee-Eung Kim (KAIST)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Spotlight: A Bayesian Approach to Generative Adversarial Imitation Learning »
Wed. Dec 5th 09:05 -- 09:10 PM Room Room 220 CD
More from the Same Authors
-
2018 Poster: Monte-Carlo Tree Search for Constrained POMDPs »
Jongmin Lee · Geon-Hyeong Kim · Pascal Poupart · Kee-Eung Kim -
2017 Poster: Generative Local Metric Learning for Kernel Regression »
Yung-Kyun Noh · Masashi Sugiyama · Kee-Eung Kim · Frank Park · Daniel Lee -
2012 Poster: Cost-Sensitive Exploration in Bayesian Reinforcement Learning »
Dongho Kim · Kee-Eung Kim · Pascal Poupart -
2012 Poster: Nonparametric Bayesian Inverse Reinforcement Learning for Multiple Reward Functions »
Jaedeug Choi · Kee-Eung Kim -
2011 Poster: MAP Inference for Bayesian Inverse Reinforcement Learning »
Jaedeug Choi · Kee-Eung Kim