Differentiable Logic Cellular Automata

Differentiable Logic Cellular Automata From Game of Life to pattern generation with learned recurrent circuits AUTHORS: Pietro Miotti Eyvind Niklasson Ettore Randazzo Alexander Mordvintsev AFFILIATIONS: Google, Paradigms of Intelligence Team PUBLISHED: March 3, 2025 Imagine trying to reverse-engineer the complex, often unexpected patterns and behaviors that emerge from simple rules. This challenge has inspired researchers and enthusiasts that work with cellular automata for decades. In cellular automata, we generally approach things from the bottom-up. We choose local rules, then investigate the resulting emergent patterns. What if we could create systems that, given some complex desired pattern, can, in a fully differentiable fashion, learn the local rules that generate it, while preserving the inherent discrete nature of cellular automata? This is what we'll explore with you today. Zoomed view of learned circuit "G" being generated by learned circuit Prior work has explored learning ... Read full article.