Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper analyzes 935 ablation experiments from 161 publications to characterize the distribution of fitness effects (DFE) in AI architectural evolution. It finds that the DFE follows a heavy-tailed Student's t-distribution, similar to biological systems like D. melanogaster and S. cerevisiae, but with a higher proportion of beneficial mutations. The study also observes logistic dynamics in architectural origination and convergent evolution of architectural traits, suggesting universal statistical laws govern evolution across substrates.
AI architecture evolution mirrors biological evolution, exhibiting a similar distribution of fitness effects and convergent trait development, suggesting universal principles govern evolutionary processes regardless of the substrate.
We test whether artificial intelligence architectural evolution obeys the same statistical laws as biological evolution. Compiling 935 ablation experiments from 161 publications, we show that the distribution of fitness effects (DFE) of architectural modifications follows a heavy-tailed Student's t-distribution with proportions (68% deleterious, 19% neutral, 13% beneficial for major ablations, n=568) that place AI between compact viral genomes and simple eukaryotes. The DFE shape matches D. melanogaster (normalized KS=0.07) and S. cerevisiae (KS=0.09); the elevated beneficial fraction (13% vs. 1-6% in biology) quantifies the advantage of directed over blind search while preserving the distributional form. Architectural origination follows logistic dynamics (R^2=0.994) with punctuated equilibria and adaptive radiation into domain niches. Fourteen architectural traits were independently invented 3-5 times, paralleling biological convergences. These results demonstrate that the statistical structure of evolution is substrate-independent, determined by fitness landscape topology rather than the mechanism of selection.