Self-Classifying MNIST Digits

Abstract

Distill articles are interactive publications and do not include traditional abstracts. This summary was written for the ML Anthology. Shows that neural cellular automata can learn to perform digit self-classification, with cells collectively determining which MNIST digit they form while exhibiting self-correcting behavior in response to perturbations.

Cite

Text

Randazzo et al. "Self-Classifying MNIST Digits." Distill, 2020. doi:10.23915/distill.00027.002

Markdown

[Randazzo et al. "Self-Classifying MNIST Digits." Distill, 2020.](https://mlanthology.org/distill/2020/randazzo2020distill-selfclassifying/) doi:10.23915/distill.00027.002

BibTeX

@article{randazzo2020distill-selfclassifying,
  title     = {{Self-Classifying MNIST Digits}},
  author    = {Randazzo, Ettore and Mordvintsev, Alexander and Niklasson, Eyvind and Levin, Michael and Greydanus, Sam},
  journal   = {Distill},
  year      = {2020},
  doi       = {10.23915/distill.00027.002},
  url       = {https://mlanthology.org/distill/2020/randazzo2020distill-selfclassifying/}
}