Google Research

Self-classifying MNIST digit CA

  • Ettore Randazzo
  • Alexander Mordvintsev {{ +moralex }
  • Eyvind Niklasson
  • Michael Levin
  • Sam Greydanus
Distill (2020)
Download Google Scholar Copy Bibtex

Abstract

Training an end-to-end differentiable, self-organising cellular automata model able to self-classify in ever-changing MNIST digits.

Research Areas

Machine Intelligence

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work

Our Research Philosophy