Fitness Landscapes and Evolvability

Smith, Tom; Husbands, Phil; Layzell, P.; O’Shea, Michael

doi:10.1162/106365602317301754

Cited by 135 publications

(100 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Knowledge of the fundamental characteristics of the problem being optimised at the level at which optimisation algorithms operate. These characteristics are represented by the fitness landscape, firstly introduced by Wright (1932), which describes the search space of an optimisation problem as a multidimensional landscape defined by the possible solutions, through which the optimisation algorithm moves, mapped to the corresponding fitness value (Smith et al, 2002). As such, the fitness landscape is not only dependent upon the problem to be solved, but also on the choice of algorithm and its parameter values.…”

Section: Overviewmentioning

confidence: 99%

“…Portraits (Smith et al, 2002) While there are numerous fitness landscape characteristics and corresponding metrics, the statistical reliability of these metrics is frequently called into question. One of the main reasons that these measures have been found to be unreliable is that no rigorous definition of the concept of "difficulty" is available in the framework of evolutionary optimisation (Kallel, 1998).…”

Section: Fitness Evolvabilitymentioning

confidence: 99%

See 1 more Smart Citation

Evolutionary algorithms and other metaheuristics in water resources: Current status, research challenges and future directions

Maier

Kapelan

Kasprzyk

et al. 2014

Environmental Modelling & Software

523

317

View full text Add to dashboard Cite

In all cases accepted manuscripts should:  link to the formal publication via its DOI  bear a CC-BY-NC-ND license -this is easy to do, click here to find out how  if aggregated with other manuscripts, for example in a repository or other site, be shared in alignment with our hosting policy  not be added to or enhanced in any way to appear more like, or to substitute for, the published journal article AbstractThe development and application of evolutionary algorithms (EAs) and other metaheuristics for the optimisation of water resources systems has been an active research field for over two decades. Research to date has emphasized algorithmic improvements and individual applications in specific areas (e.g., model calibration, water distribution systems, groundwater management, river-basin planning and management, etc.). However, there has been limited synthesis between shared problem traits, common EA challenges, and needed advances across major applications. This paper clarifies the current status and future research directions for better solving key water resources problems using EAs. Advances in understanding fitness landscape properties and their effects on algorithm performance are critical. Future EA-based applications to real-world problems require a fundamental shift of focus towards improving problem formulations, understanding general theoretic frameworks for problem decompositions, major advances in EA computational efficiency, and most importantly aiding real decision-making in complex, uncertain application contexts.

show abstract

Section: Overviewmentioning

confidence: 99%

Section: Fitness Evolvabilitymentioning

confidence: 99%

Evolutionary algorithms and other metaheuristics in water resources: Current status, research challenges and future directions

Maier

Kapelan

Kasprzyk

et al. 2014

Environmental Modelling & Software

523

317

View full text Add to dashboard Cite

show abstract

“…Evolvability is the ability of a population to produce offspring fitter than any yet in existence [1], and not to produce less fit variants [13], and is therefore fundamental to the process of evolution itself. Evolvability is also known as evolutionary adaptability [8] and as such, a major element of evolvability is the capacity to adapt to changing environments by learning to exploit commonalities over time in those environments.…”

Section: Introductionmentioning

confidence: 99%

Analysing the Evolvability of Neural Network Agents Through Structural Mutations

Schlessinger

Bentley

Lotto

2005

Advances in Artificial Life

View full text Add to dashboard Cite

Abstract. This paper investigates evolvability of artificial neural networks within an artificial life environment. Five different structural mutations are investigated, including adaptive evolution, structure duplication, and incremental changes. The total evolvability indicator, E total , and the evolvability function through time, are calculated in each instance, in addition to other functional attributes of the system. The results indicate that incremental modifications to networks, and incorporating an adaptive element into the evolution process itself, significantly increases neural network evolvability within open-ended artificial life simulations.

show abstract

“…" " (arms race) [3] 1 [4] [4,5] [6] (ruggedness) (neutrality) [3,[7][8][9][10][11] Ebner [12] ---25 - …”

mentioning

confidence: 99%

Tracking the Red Queen Effect by Estimating Features of Competitive Co-Evolutionary Fitness Landscapes

Katada

Handa

2013

Transactions of ISCIE

View full text Add to dashboard Cite

Open-ended evolution is considered to be caused by several factors, one of which would be coevolution. Competitive co-evolution can give rise to the "Red Queen effect", where the fitness landscape of each population is continuously changed by the competing population. Therefore, if such continuous changes are captured, co-evolutionary progress would be measured. In this paper, we estimate features of competitive co-evolutionary fitness landscapes on a predator-prey problem in computer simulations and investigate the Red Queen effect on the fitness landscape. Two types of method were proposed to estimate features, ruggedness and neutrality. One was calculated based on accumulated data so far at each generation, and the other was based on accumulated data during a certain period. The results suggest to us that our method can track the progress of fitness landscapes on competitive co-evolutionary robotics.

show abstract

Fitness Landscapes and Evolvability

Cited by 135 publications

References 30 publications

Evolutionary algorithms and other metaheuristics in water resources: Current status, research challenges and future directions

Evolutionary algorithms and other metaheuristics in water resources: Current status, research challenges and future directions

Analysing the Evolvability of Neural Network Agents Through Structural Mutations

Tracking the Red Queen Effect by Estimating Features of Competitive Co-Evolutionary Fitness Landscapes

Contact Info

Product

Resources

About