Skip to yearly menu bar Skip to main content


EMI: Exploration with Mutual Information

Hyoungseok Kim · Jaekyeom Kim · Yeonwoo Jeong · Sergey Levine · Hyun Oh Song

Pacific Ballroom #46

Keywords: [ Deep Reinforcement Learning ]


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 competitive results on challenging locomotion tasks with continuous control and on image-based exploration tasks with discrete actions on Atari. The source code is available at

Live content is unavailable. Log in and register to view live content