papers AI Learner
The Github is limit! Click to go to the new site.

Gradient-free Policy Architecture Search and Adaptation

2017-10-16
Sayna Ebrahimi, Anna Rohrbach, Trevor Darrell

Abstract

We develop a method for policy architecture search and adaptation via gradient-free optimization which can learn to perform autonomous driving tasks. By learning from both demonstration and environmental reward we develop a model that can learn with relatively few early catastrophic failures. We first learn an architecture of appropriate complexity to perceive aspects of world state relevant to the expert demonstration, and then mitigate the effect of domain-shift during deployment by adapting a policy demonstrated in a source domain to rewards obtained in a target environment. We show that our approach allows safer learning than baseline methods, offering a reduced cumulative crash metric over the agent’s lifetime as it learns to drive in a realistic simulated environment.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1710.05958

PDF

https://arxiv.org/pdf/1710.05958


Similar Posts

Comments