Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Diffusion Models are Evolutionary Algorithms

About

In a convergence of machine learning and biology, we reveal that diffusion models are evolutionary algorithms. By considering evolution as a denoising process and reversed evolution as diffusion, we mathematically demonstrate that diffusion models inherently perform evolutionary algorithms, naturally encompassing selection, mutation, and reproductive isolation. Building on this equivalence, we propose the Diffusion Evolution method: an evolutionary algorithm utilizing iterative denoising -- as originally introduced in the context of diffusion models -- to heuristically refine solutions in parameter spaces. Unlike traditional approaches, Diffusion Evolution efficiently identifies multiple optimal solutions and outperforms prominent mainstream evolutionary algorithms. Furthermore, leveraging advanced concepts from diffusion models, namely latent space diffusion and accelerated sampling, we introduce Latent Space Diffusion Evolution, which finds solutions for evolutionary tasks in high-dimensional complex parameter space while significantly reducing computational steps. This parallel between diffusion and evolution not only bridges two different fields but also opens new avenues for mutual enhancement, raising questions about open-ended evolution and potentially utilizing non-Gaussian or discrete diffusion models in the context of Diffusion Evolution.

Yanbo Zhang, Benedikt Hartl, Hananel Hazan, Michael Levin• 2024

Related benchmarks

TaskDatasetResultRank
OptimizationBeale
Average Entropy S5.5
12
Fitness Landscape OptimizationRastrigin 4 dimensions
Average Entropy5.99
6
Fitness Landscape OptimizationRastrigin 32 dimensions
Average Entropy6
6
Fitness Landscape OptimizationRastrigin 256 dimensions
Average Entropy6
6
Quality-Diversity OptimizationRastrigin 4 dimensions
QD Score33.1
6
Quality-Diversity OptimizationRastrigin 32 dimensions
QD-score73.4
6
Quality-Diversity OptimizationRastrigin 256 dimensions
QD-score70.2
6
Fitness Landscape OptimizationRosenbrock
Average Entropy5.86
6
Fitness Landscape OptimizationHimmelblau
Average Entropy5.23
6
Fitness Landscape OptimizationAckley
Average Entropy5.67
6
Showing 10 of 16 rows

Other info

Follow for update