Imagination-Augmented Agents for Deep Reinforcement Learning
Machine Learning
2018-02-15 v2 Artificial Intelligence
Machine Learning
Abstract
We introduce Imagination-Augmented Agents (I2As), a novel architecture for deep reinforcement learning combining model-free and model-based aspects. In contrast to most existing model-based reinforcement learning and planning methods, which prescribe how a model should be used to arrive at a policy, I2As learn to interpret predictions from a learned environment model to construct implicit plans in arbitrary ways, by using the predictions as additional context in deep policy networks. I2As show improved data efficiency, performance, and robustness to model misspecification compared to several baselines.
Cite
@article{arxiv.1707.06203,
title = {Imagination-Augmented Agents for Deep Reinforcement Learning},
author = {Théophane Weber and Sébastien Racanière and David P. Reichert and Lars Buesing and Arthur Guez and Danilo Jimenez Rezende and Adria Puigdomènech Badia and Oriol Vinyals and Nicolas Heess and Yujia Li and Razvan Pascanu and Peter Battaglia and Demis Hassabis and David Silver and Daan Wierstra},
journal= {arXiv preprint arXiv:1707.06203},
year = {2018}
}