Reinforcement Learning Generalization with Surprise Minimization

Jerry Zikun Chen

Reinforcement Learning Generalization with Surprise Minimization

Machine Learning 2020-07-06 v2 Artificial Intelligence

Authors: Jerry Zikun Chen

Abstract

Generalization remains a challenging problem for deep reinforcement learning algorithms, which are often trained and tested on the same set of deterministic game environments. When test environments are unseen and perturbed but the nature of the task remains the same, generalization gaps can arise. In this work, we propose and evaluate a surprise minimizing agent on a generalization benchmark to show an additional reward learned from a simple density model can show robustness in procedurally generated game environments that provide constant source of entropy and stochasticity.

Keywords

generalization in machine learning reinforcement learning regularization

Cite

@article{arxiv.2004.12399,
  title  = {Reinforcement Learning Generalization with Surprise Minimization},
  author = {Jerry Zikun Chen},
  journal= {arXiv preprint arXiv:2004.12399},
  year   = {2020}
}

Comments

Inductive biases, invariances and generalization in RL Workshop, ICML 2020

Related papers

View all related →