ML Anthology

Understanding RL Vision

Jacob Hilton, Nick Cammarata, Shan Carter, Gabriel Goh, Chris Olah
Distill 2020
doi:10.23915/distill.00029 /distill/2020/hilton2020distill-understanding/

Abstract

Distill articles are interactive publications and do not include traditional abstracts. This summary was written for the ML Anthology. Applies interpretability techniques to a reinforcement learning agent trained on the CoinRun game, using attribution and dimensionality reduction to identify which game objects influence decisions, diagnose failures, and perform targeted model editing.

Distill Code Semantic Scholar

Cite

Text

Hilton et al. "Understanding RL Vision." Distill, 2020. doi:10.23915/distill.00029

Markdown

[Hilton et al. "Understanding RL Vision." Distill, 2020.](https://mlanthology.org/distill/2020/hilton2020distill-understanding/) doi:10.23915/distill.00029

BibTeX

@article{hilton2020distill-understanding,
  title     = {{Understanding RL Vision}},
  author    = {Hilton, Jacob and Cammarata, Nick and Carter, Shan and Goh, Gabriel and Olah, Chris},
  journal   = {Distill},
  year      = {2020},
  doi       = {10.23915/distill.00029},
  url       = {https://mlanthology.org/distill/2020/hilton2020distill-understanding/}
}