ML Anthology

Naturally Occurring Equivariance in Neural Networks

Chris Olah, Nick Cammarata, Chelsea Voss, Ludwig Schubert, Gabriel Goh
Distill 2020
doi:10.23915/distill.00024.004 /distill/2020/olah2020distill-naturally/

Abstract

Distill articles are interactive publications and do not include traditional abstracts. This summary was written for the ML Anthology. Demonstrates how neural networks naturally learn multiple transformed copies of the same feature connected by symmetric weights, showing that equivariance emerges organically in trained networks without being explicitly designed.

Distill Code Semantic Scholar

Cite

Text

Olah et al. "Naturally Occurring Equivariance in Neural Networks." Distill, 2020. doi:10.23915/distill.00024.004

Markdown

[Olah et al. "Naturally Occurring Equivariance in Neural Networks." Distill, 2020.](https://mlanthology.org/distill/2020/olah2020distill-naturally/) doi:10.23915/distill.00024.004

BibTeX

@article{olah2020distill-naturally,
  title     = {{Naturally Occurring Equivariance in Neural Networks}},
  author    = {Olah, Chris and Cammarata, Nick and Voss, Chelsea and Schubert, Ludwig and Goh, Gabriel},
  journal   = {Distill},
  year      = {2020},
  doi       = {10.23915/distill.00024.004},
  url       = {https://mlanthology.org/distill/2020/olah2020distill-naturally/}
}