Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Coralai: Intrinsic Evolution of Embodied Neural Cellular Automata Ecosystems

Aidan Barbieux, Rodrigo Canaan

TL;DR

Coralai is presented, a framework for exploring diverse ecosystems of Neural Cellular Automata (NCA) using modular, GPU-accelerated Taichi kernels to interact, enact environmental changes, and evolve through local survival, merging, and mutation operations implemented with HyperNEAT and PyTorch.

Abstract

This paper presents Coralai, a framework for exploring diverse ecosystems of Neural Cellular Automata (NCA). Organisms in Coralai utilize modular, GPU-accelerated Taichi kernels to interact, enact environmental changes, and evolve through local survival, merging, and mutation operations implemented with HyperNEAT and PyTorch. We provide an exploratory experiment implementing physics inspired by slime mold behavior showcasing the emergence of competition between sessile and mobile organisms, cycles of resource depletion and recovery, and symbiosis between diverse organisms. We conclude by outlining future work to discover simulation parameters through measures of multi-scale complexity and diversity. Code for Coralai is available at https://github.com/aidanbx/coralai , video demos are available at https://www.youtube.com/watch?v=NL8IZQY02-8 .

Coralai: Intrinsic Evolution of Embodied Neural Cellular Automata Ecosystems

TL;DR

Coralai is presented, a framework for exploring diverse ecosystems of Neural Cellular Automata (NCA) using modular, GPU-accelerated Taichi kernels to interact, enact environmental changes, and evolve through local survival, merging, and mutation operations implemented with HyperNEAT and PyTorch.

Abstract

This paper presents Coralai, a framework for exploring diverse ecosystems of Neural Cellular Automata (NCA). Organisms in Coralai utilize modular, GPU-accelerated Taichi kernels to interact, enact environmental changes, and evolve through local survival, merging, and mutation operations implemented with HyperNEAT and PyTorch. We provide an exploratory experiment implementing physics inspired by slime mold behavior showcasing the emergence of competition between sessile and mobile organisms, cycles of resource depletion and recovery, and symbiosis between diverse organisms. We conclude by outlining future work to discover simulation parameters through measures of multi-scale complexity and diversity. Code for Coralai is available at https://github.com/aidanbx/coralai , video demos are available at https://www.youtube.com/watch?v=NL8IZQY02-8 .
Paper Structure (4 sections, 2 figures)

This paper contains 4 sections, 2 figures.

Table of Contents

  1. Introduction
  2. System Design
  3. Experiment
  4. Discussion and Future Work

Figures (2)

  • Figure 1: A single update in Coralai. (A): The sensing of substrate and production of actions. (B): Applying the actions of the entire system to the substrate via 'physics'. (C): Visualization and radiation of the substrate resulting in mutation of CPPN genomes and generation of dense network parameters.
  • Figure 2: Notable experiment runs. Visual observation shows competition between mobile and sessile organisms in (A), extinction and replacement in (B), a variety of organisms stabilized in an ecosystem in (C), and competing slime-mold-like organisms in (D). Energy is displayed as red, infrastructure by green, and genome/species by blue.