skip to main content


Title: Interaction Between Evolution and Learning in NK Fitness Landscapes
Artificial Life has a long tradition of studying the interaction between learning and evolution. And, thanks to the increase in the use of individual learning techniques in Artificial Intelligence, there has been a recent revival of work combining individual and evolutionary learning. Despite the breadth of work in this area, the exact trade-offs between these two forms of learning remain unclear. In this work, we systematically examine the effect of task difficulty, the individual learning approach, and the form of inheritance on the performance of the population across different combinations of learning and evolution. We analyze in depth the conditions in which hybrid strategies that combine lifetime and evolutionary learning outperform either lifetime or evolutionary learning in isolation. We also discuss the importance of these results in both a biological and algorithmic context.  more » « less
Award ID(s):
1845322
NSF-PAR ID:
10174175
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
ALIFE 2020: The 2020 Conference on Artificial Life
Issue:
32
Page Range / eLocation ID:
761 - 767
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Living organisms learn on multiple time scales: evolutionary as well as individual-lifetime learning. These two learning modes are complementary: the innate phenotypes developed through evolution significantly influence lifetime learning. However, it is still unclear how these two learning methods interact and whether there is a benefit to part of the system being optimized on a different time scale using a population-based approach while the rest of it is trained on a different time-scale using an individualistic learning algorithm. In this work, we study the benefits of such a hybrid approach using an actor-critic framework where the critic part of an agent is optimized over evolutionary time based on its ability to train the actor part of an agent during its lifetime. Typically, critics are optimized on the same time-scale as the actor using the Bellman equation to represent long-term expected reward. We show that evolution can find a variety of different solutions that can still enable an actor to learn to perform a behavior during its lifetime. We also show that although the solutions found by evolution represent different functions, they all provide similar training signals during the lifetime. This suggests that learning on multiple time-scales can effectively simplify the overall optimization process in the actor-critic framework by finding one of many solutions that can still train an actor just as well. Furthermore, analysis of the evolved critics can yield additional possibilities for reinforcement learning beyond the Bellman equation. 
    more » « less
  2. Abstract

    The fundamental tradeoff between carbon gain and water loss has long been predicted as an evolutionary driver of plant strategies across environments. Nonetheless, challenges in measuring carbon gain and water loss in ways that integrate over leaf lifetime have limited our understanding of the variation in and mechanistic bases of this tradeoff. Furthermore, the microevolution of plant traits within species versus the macroevolution of strategies among closely related species may not be the same, and accordingly, the latter must be addressed using comparative phylogenetic analyses.

    Here we introduce the concept of ‘integrated metabolic strategy’ (IMS) to describe the ratio between carbon isotope composition (δ13C) and oxygen isotope composition above source water (Δ18O) of leaf cellulose. IMS is a measure of leaf‐level conditions that integrate several mechanisms contributing to carbon gain (δ13C) and water loss (Δ18O) over leaf lifespan, with larger values reflecting higher metabolic efficiency and hence less of a tradeoff. We tested how IMS evolves among closely related yet ecologically diverse milkweed species, and subsequently addressed phenotypic plasticity in response to water availability in species with divergent IMS.

    Integrated metabolic strategy varied strongly among 20Asclepiasspecies when grown under controlled conditions, and phylogenetic analyses demonstrate species‐specific tradeoffs between carbon gain and water loss. Larger IMS values were associated with species from dry habitats, with larger carboxylation capacity, smaller stomatal conductance and smaller leaves; smaller IMS was associated with wet habitats, smaller carboxylation capacity, larger stomatal conductance and larger leaves. The evolution of IMS was dominated by changes in species’ demand for carbon (δ13C) more so than water conservation (Δ18O). Although some individual physiological traits showed phylogenetic signal, IMS did not.

    In response to experimental decreases in soil moisture, three species maintained similar IMS across levels of water availability because of proportional increases inδ13C and Δ18O (or little change in either), while one species increased IMS due to disproportional changes inδ13C relative to Δ18O.

    Synthesis.IMS is a broadly applicable mechanistic tool; IMS variation among and within species may shed light on unresolved questions relating to the evolution and ecology of plant ecophysiological strategies.

     
    more » « less
  3. Humans have long known how to co-opt evolutionary processes for their own benefit. Carefully choosing which individuals to breed so that beneficial traits would take hold, they have domesticated dogs, wheat, cows and many other species to fulfil their needs. Biologists have recently refined these ‘artificial selection’ approaches to focus on microorganisms. The hope is to obtain microbes equipped with desirable features, such as the ability to degrade plastic or to produce valuable molecules. However, existing ways of using artificial selection on microbes are limited and sometimes not effective. Computer scientists have also harnessed evolutionary principles for their own purposes, developing highly effective artificial selection protocols that are used to find solutions to challenging computational problems. Yet because of limited communication between the two fields, sophisticated selection protocols honed over decades in evolutionary computing have yet to be evaluated for use in biological populations. In their work, Lalejini et al. compared popular artificial selection protocols developed for either evolutionary computing or work with microorganisms. Two computing selection methods showed promise for improving directed evolution in the laboratory. Crucially, these selection protocols differed from conventionally used methods by selecting for both diversity and performance, rather than performance alone. These promising approaches are now being tested in the laboratory, with potentially far-reaching benefits for medical, biotech, and agricultural applications. While evolutionary computing owes its origins to our understanding of biological processes, it has much to offer in return to help us harness those same mechanisms. The results by Lalejini et al. help to bridge the gap between computational and biological communities who could both benefit from increased collaboration. 
    more » « less
  4. Abstract

    The rich diversity of angiosperms, both the planet’s dominant flora and the cornerstone of agriculture, is integrally intertwined with a distinctive evolutionary history. Here, we explore the interplay between angiosperm genome organization and botanical diversity, empowered by genomic approaches ranging from genetic mapping to analysis of gene regulation. Commonality in the genetic hardware of plants has enabled robust comparative genomics that has provided a broad picture of angiosperm evolution and implicated both general processes and specific elements in contributing to botanical diversity. We argue that the hardware of plant genomes – both in content and in dynamics – has been shaped by selection for rather substantial differences in gene regulation between plants and animals such as maize and human, organisms of comparable genome size and gene number. Their distinctive genome content and dynamics may reflect in part the indeterminate development of plants that puts strikingly different demands on gene regulation than in animals. Repeated polyploidization of plant genomes and multiplication of individual genes together with extensive rearrangement and differential retention, provides rich raw material for selection of morphological and/or physiological variations conferring fitness in specific niches, whether natural or artificial. These findings exemplify the burgeoning information available to employ in increasing knowledge of plant biology, and in modifying selected plants to better meet human needs.

     
    more » « less
  5. Embryonic development is arguably the most complex process an organism undergoes during its lifetime, and understanding this complexity is best approached with a systems-level perspective. The sea urchin has become a highly valuable model organism for understanding developmental specification, morphogenesis, and evolution. As a non-chordate deuterostome, the sea urchin occupies an important evolutionary niche between protostomes and vertebrates. Lytechinus variegatus (Lv) is an Atlantic species that has been well studied, and which has provided important insights into signal transduction, patterning, and morphogenetic changes during embryonic and larval development. The Pacific species, Strongylocentrotus purpuratus (Sp), is another well-studied sea urchin, particularly for gene regulatory networks (GRNs) and cis-regulatory analyses. A well-annotated genome and transcriptome for Sp are available, but similar resources have not been developed for Lv. Here, we provide an analysis of the Lv transcriptome at 11 timepoints during embryonic and larval development. Temporal analysis suggests that the gene regulatory networks that underlie specification are well-conserved among sea urchin species. We show that the major transitions in variation of embryonic transcription divide the developmental time series into four distinct, temporally sequential phases. Our work shows that sea urchin development occurs via sequential intervals of relatively stable gene expression states that are punctuated by abrupt transitions. 
    more » « less