English

Dex: Incremental Learning for Complex Environments in Deep Reinforcement Learning

Machine Learning 2017-06-20 v1 Artificial Intelligence Machine Learning

Abstract

This paper introduces Dex, a reinforcement learning environment toolkit specialized for training and evaluation of continual learning methods as well as general reinforcement learning problems. We also present the novel continual learning method of incremental learning, where a challenging environment is solved using optimal weight initialization learned from first solving a similar easier environment. We show that incremental learning can produce vastly superior results than standard methods by providing a strong baseline method across ten Dex environments. We finally develop a saliency method for qualitative analysis of reinforcement learning, which shows the impact incremental learning has on network attention.

Keywords

Cite

@article{arxiv.1706.05749,
  title  = {Dex: Incremental Learning for Complex Environments in Deep Reinforcement Learning},
  author = {Nick Erickson and Qi Zhao},
  journal= {arXiv preprint arXiv:1706.05749},
  year   = {2017}
}

Comments

NIPS 2017 submission, 10 pages, 26 figures

R2 v1 2026-06-22T20:22:14.566Z