Selecting for evolvable representations

Reisinger, Joseph; Miikkulainen, Risto

doi:10.1145/1143997.1144199

Cited by 14 publications

(29 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Previous studies have indicated that temporally varying environments can affect several properties of evolved systems such as their structure (6), robustness (21), evolvability (22,23), and genotype-phenotype mapping (10,24). In particular, goals that change over time in a modular fashion (18), such that each new goal shares some of the subproblems with the previous goal, were found to spontaneously generate systems with modular structure (18).…”

mentioning

confidence: 99%

Varying environments can speed up evolution

Kashtan

Noor

Alon

2007

Proc. Natl. Acad. Sci. U.S.A.

267

272

View full text Add to dashboard Cite

Simulations of biological evolution, in which computers are used to evolve systems toward a goal, often require many generations to achieve even simple goals. It is therefore of interest to look for generic ways, compatible with natural conditions, in which evolution in simulations can be speeded. Here, we study the impact of temporally varying goals on the speed of evolution, defined as the number of generations needed for an initially random population to achieve a given goal. Using computer simulations, we find that evolution toward goals that change over time can, in certain cases, dramatically speed up evolution compared with evolution toward a fixed goal. The highest speedup is found under modularly varying goals, in which goals change over time such that each new goal shares some of the subproblems with the previous goal. The speedup increases with the complexity of the goal: the harder the problem, the larger the speedup. Modularly varying goals seem to push populations away from local fitness maxima, and guide them toward evolvable and modular solutions. This study suggests that varying environments might significantly contribute to the speed of natural evolution. In addition, it suggests a way to accelerate optimization algorithms and improve evolutionary approaches in engineering.biological physics ͉ modularity ͉ optimization ͉ systems biology A central question is how evolution can explain the speed at which the present complexity of life arose (1-17). Current computer simulations of evolution are well known to have difficulty in scaling to high complexity. Such studies use computers to evolve artificial systems, which serve as an analogy to biological systems, toward a given goal (6, 9, 18). The simulations mimic natural evolution by incorporating replication, variation (e.g., mutation and recombination), and selection. Typically, a logarithmic slowdown in evolution is observed: longer and longer periods are required for successive improvements in fitness (6,9,18) [similar slowdown is observed in adaptation experiments on bacteria in constant environments (19,20)]. Simulations can take many thousands of generations to reach even relatively simple goals, such as Boolean functions of several variables (9, 18). Thus, to understand the speed of natural evolution, it is of interest to find generic ways, compatible with natural conditions, in which evolution in simulations can be speeded.To address this, we consider here the fact that the environment of organisms in nature changes over time. Previous studies have indicated that temporally varying environments can affect several properties of evolved systems such as their structure (6), robustness (21), evolvability (22, 23), and genotype-phenotype mapping (10, 24). In particular, goals that change over time in a modular fashion (18), such that each new goal shares some of the subproblems with the previous goal, were found to spontaneously generate systems with modular structure (18).Here, we study the effect of temporally varying environments on the speed...

show abstract

mentioning

confidence: 99%

Varying environments can speed up evolution

Kashtan

Noor

Alon

2007

Proc. Natl. Acad. Sci. U.S.A.

267

272

View full text Add to dashboard Cite

show abstract

“…Such motif significance profiles are typical of neural networks found in nature [13]. Such highly recurrent network structures may be used to increase the detectability of mutations affecting evolvability [18].…”

Section: Network Motifsmentioning

confidence: 89%

“…One possible reason is that with the implicit encoding, evolution controls how the phenotypes are distributed, thereby transforming random mutations into structured phenotypic variation. Furthermore, since such mutations have an immediate and detectable impact on fitness, fewer mutations are required on average to create a gradient for selection [18].…”

Section: Discussion and Future Workmentioning

confidence: 99%

Acquiring evolvability through adaptive representations

Reisinger

Miikkulainen

2007

Proceedings of the 9th Annual Conference on Genetic and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

Adaptive representations allow evolution to explore the space of phenotypes by choosing the most suitable set of genotypic parameters. Although such an approach is believed to be efficient on complex problems, few empirical studies have been conducted in such domains. In this paper, three neural network representations, a direct encoding, a complexifying encoding, and an implicit encoding capable of adapting the genotype-phenotype mapping are compared on Nothello, a complex game playing domain from the AAAI General Game Playing Competition. Implicit encoding makes the search more efficient and uses several times fewer parameters. Random mutation leads to highly structured phenotypic variation that is acquired during the course of evolution rather than built into the representation itself. Thus, adaptive representations learn to become evolvable, and furthermore do so in a way that makes search efficient on difficult coevolutionary problems.

show abstract

“…Thus, high evolvability can be detected and favored by such selection pressure. For an artificial evolutionary system Reisinger et al concur when they propose an indirect encoding representation to improve evolvability [56,57]. A gradually changing fitness function Journal of Artificial Evolution and Applications 3 is designed to measure evolvability of representations and to evolve a population that is adaptive under different environments.…”

Section: Definitionmentioning

confidence: 99%

“…They are sensitive to training data and to varying constraints of different models; so a common framework would be required. Moreover, Reisinger and Miikkulainen [56] propose an evolvable representation and an evaluation strategy to exert indirect selection pressure on evolvability. In their work, a systematically changing fitness function is adopted according to a special evolvable representation that can reflect efficiently how genetic changes restructure phenotypic variation.…”

Section: Definitionmentioning

confidence: 99%

Evolvability and Speed of Evolutionary Algorithms in Light of Recent Developments in Biology

Banzhaf

2010

Journal of Artificial Evolution and Applications

View full text Add to dashboard Cite

Biological and artificial evolutionary systems exhibit varying degrees of evolvability and different rates of evolution. Such quantities can be affected by various factors. Here, we review some evolutionary mechanisms and discuss new developments in biology that can potentially improve evolvability or accelerate evolution in artificial systems. Biological notions are discussed to the degree they correspond to notions in Evolutionary Computation. We hope that the findings put forward here can be used to design computational models of evolution that produce significant gains in evolvability and evolutionary speed.

show abstract

Selecting for evolvable representations

Cited by 14 publications

References 27 publications

Varying environments can speed up evolution

Varying environments can speed up evolution

Acquiring evolvability through adaptive representations

Evolvability and Speed of Evolutionary Algorithms in Light of Recent Developments in Biology

Contact Info

Product

Resources

About