John Conway was a particularly prolific and curious mathematician.

Geometry, number theory, knots, he left an indelible imprint on so many parts of the math world.

We have to make sure we say that first, before we get to the thing that brought him a lot of fame and recognition outside of the math world,

something called the Game of Life.

This relatively simple computer simulation more or less went viral in the early 1970s,

and you might know it.

You start with a grid, and every cell on that grid can be either empty or full, dead or alive.

And it ticks forward one step at a time and everything stays the same unless certain conditions are met.

A dead cell, for example, will come alive if there are three alive cells around it.

A living cell that has less than two or more than three living cells around it. A living cell that has less than two or more than three living

cells around it dies on the next step. And that's it. From those simple rules and different

starting configurations of living cells, you can develop a wide range of behaviors and patterns.

It's the most famous example of what are called cellular automata,

and it serves as an easy way to demonstrate how complexity and diversity can come from a few simple rules.

But it can be more than a metaphor too, and there's an emerging field linking cellular

automata with neural networks in ways that could have us rethinking how we structure computing.

Welcome to the Quantum Podcast, where we explore the frontiers of fundamental science and math.

I'm Samir Patel, editor-in-chief of Quanta magazine.

Science journalist and Quanta contributing writer George Musser recently covered some of this

new work, which is adding layers and layers of modern computing complexity on the old idea of cellular automata

to come up with something new. His story is called self-assembly gets automated in reverse of

game of life, and he's here to speak with us about it today. Welcome back to the show, George.

...