Implications From the Analogous Relationship Between Evolutionary and Learning Processes

Abstract

Organismal evolution is a process of discovering better-fitting phenotypes through trial and error across generations. This iterative process resembles learning processes, an analogy recognized since the 1950s. Recognizing this parallel suggests that evolutionary biology and machine learning can mutually benefit from each other; however, ample opportunities for research into their corresponding concepts remain. In this review, we aim to enhance predictive capabilities and theoretical developments in both fields by exploring their conceptual parallels through specific examples that have emerged from recent advances. We focus on the importance of moving beyond predictions by machine learning approaches for specific cases, but instead advocate for interpretable machine learning approaches for discovering common laws for predicting evolutionary outcomes. This approach seeks to establish a theoretical framework that can transform evolutionary science into a field enriched with predictive theory while also inspiring new modeling and algorithmic strategies in machine learning.