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

Shvartzman, Boaz and Ram, Yoav and Kouyos, Roger Dimitri (2024) Self-replicating artificial neural networks give rise to universal evolutionary dynamics. PLOS Computational Biology, 20 (3). e1012004. ISSN 1553-7358

[thumbnail of journal.pcbi.1012004.pdf] Text
journal.pcbi.1012004.pdf - Published Version

Download (2MB)

Abstract

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.

Item Type: Article
Subjects: GO for STM > Biological Science
Depositing User: Unnamed user with email support@goforstm.com
Date Deposited: 10 Apr 2024 11:21
Last Modified: 10 Apr 2024 11:21
URI: http://archive.article4submit.com/id/eprint/2769

Actions (login required)

View Item
View Item