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

EMI: Exploration with Mutual Information

2019-01-24
Hyoungseok Kim, Jaekyeom Kim, Yeonwoo Jeong, Sergey Levine, Hyun Oh Song

Abstract

Reinforcement learning algorithms struggle when the reward signal is very sparse. In these cases, naive random exploration methods essentially rely on a random walk to stumble onto a rewarding state. Recent works utilize intrinsic motivation to guide the exploration via generative models, predictive forward models, or discriminative modeling of novelty. We propose EMI, which is an exploration method that constructs embedding representation of states and actions that does not rely on generative decoding of the full observation but extracts predictive signals that can be used to guide exploration based on forward prediction in the representation space. Our experiments show that the proposed method significantly outperforms a number of existing exploration methods on challenging locomotion task with continuous control and on image-based exploration tasks with discrete actions on Atari.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1810.01176

PDF

http://arxiv.org/pdf/1810.01176


Similar Posts

Comments