Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales · HackerPapers discussions

Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales — mentioned by DonHopkins

Mon, 27 Jul 2026 09:24:39 GMT

Good point! To respect EA's trademark and the terms of their modifications to the GPL-3 license regarding the use of their SimCity trademark, I renamed the open source version of SimCity to Micropolis, which was its original working title. It would be safe to call this project "PGMicropolis" as long as it wasn't for profit. At the time, there was a disk drive manufacturer by that name, so Will Wright renamed it SimCity. Fortunately the guy who now owns the rights to the Micropolis trademark (and is an old school hacker who really gets it) was kind enough to allow us to use the name for non…

Paper and discussion history. A mention is not an endorsement.

Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales — mentioned by DonHopkins

Wed, 11 Feb 2026 21:12:21 GMT

Sam Earle's fractal neural network approach (https://arxiv.org/pdf/2002.03896) is exactly this kind of thing. His key trick: recursive weight-sharing in fractal convolutional blocks, so each column of the network acts as a continuous-valued cellular automaton ticking different numbers of times. The deepest column gets a 33x33 receptive field -- enough to connect power across an entire 32x32 map in one forward pass. The agents discovered power-plant + residential pairing, road placement for density, zone clustering by type, and traffic-avoiding road layouts. When stuck at local optima, a human…

Paper and discussion history. A mention is not an endorsement.

Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales — mentioned by DonHopkins

Wed, 11 Feb 2026 21:09:39 GMT

I developed the open source version of this game, called Micropolis. Great to see more people building on it! A few years before the LLM era, Sam Earle took a different approach -- training reinforcement learning agents with fractal neural networks to play Micropolis, optimizing for population at variable map scales: Using Fractal Neural Networks to Play SimCity 1 and Conway’s Game of Life at Variable Scales: https://arxiv.org/pdf/2002.03896 His gym-city repo wraps Micropolis as an OpenAI Gym environment: https://github.com/smearle/gym-city The interesting finding was that fractal…

Paper and discussion history. A mention is not an endorsement.