Sex with no regrets: How sexual reproduction uses a no regret learning algorithm for evolutionary advantage. (7th August 2017)

• Record Type: Journal Article
• Authors: Edhan, Omer; Hellman, Ziv; Sherill-Rofe, Dana
• Is Part Of: Journal of theoretical biology. Volume 426(2017)
• Page Start: 67
• Page End: 81
• Publication Date: 2017-08-07
• Subjects: Evolution -- Sexual reproduction -- Learning algorithms; Biology -- Periodicals; Biological Science Disciplines -- Periodicals; Biology -- Periodicals; Biologie -- Périodiques; Theoretische biologie; Biology; Periodicals; 571.05
• DOI: 10.1016/j.jtbi.2017.05.018
• Languages: English
• ISSNs: 0022-5193

Abstract

Highlights: Sex implements a machine learning algorithm that more often than not grants it an evolutionary advantage over asexual reproduction. The algorithm reliably works even though each generation is a tiny sample of the full space of all possible genotypes. The sparseness of each generation's genotypes in the space of all genotypes is crucial for sex to gain an advantage. Computer simulations indicate that this is sufficient to give sex an evolutionary edge, even in stable and unchanging environments. Asexual populations rapidly reach a fitness plateau, but is eventually surpassed, with high probability, by the fitness levels of sexual populations. Abstract: The question of 'why sex' has long been a puzzle. The randomness of recombination, which potentially produces low fitness progeny, contradicts notions of fitness landscape hill climbing. We use the concept of evolution as an algorithm for learning unpredictable environments to provide a possible answer. While sex and asex both implement similar machine learning no-regret algorithms in the context of random samples that are small relative to a vast genotype space, the algorithm of sex constitutes a more efficient goal-directed walk through this space. Simulations indicate this gives sex an evolutionary advantage, even in stable, unchanging environments. Asexual populations rapidly reach a fitness plateau, but the learning aspect of the no-regret algorithm most often eventually boosts the fitness of sexual populations past the maximal viability of corresponding asexual populations. In this light, the randomness of sexual recombination is not a hindrance but a crucial component of the 'sampling for learning' algorithm of sexual reproduction.