Self-replicating artificial neural networks give rise to universal evolutionary dynamics

Boaz Shvartzman, Yoav Ram*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


In evolutionary models, mutations are exogenously introduced by the modeler, rather than endogenously introduced by the replicator itself. We present a new deep-learning based computational model, the self-replicating artificial neural network (SeRANN). We train it to (i) copy its own genotype, like a biological organism, which introduces endogenous spontaneous mutations; and (ii) simultaneously perform a classification task that determines its fertility. Evolving 1,000 SeRANNs for 6,000 generations, we observed various evolutionary phenomena such as adaptation, clonal interference, epistasis, and evolution of both the mutation rate and the distribution of fitness effects of new mutations. Our results demonstrate that universal evolutionary phenomena can naturally emerge in a self-replicator model when both selection and mutation are implicit and endogenous. We therefore suggest that SeRANN can be applied to explore and test various evolutionary dynamics and hypotheses.

Original languageEnglish
Article numbere1012004
JournalPLoS Computational Biology
Issue number3
StatePublished - Mar 2024


FundersFunder number
NVIDIA Accelerated Data Science GPU
Australian Wildlife Society
Minerva Stiftung Center for Lab Evolution
Israel Science Foundation552/19
Israel Science Foundation


    Dive into the research topics of 'Self-replicating artificial neural networks give rise to universal evolutionary dynamics'. Together they form a unique fingerprint.

    Cite this