Performance of Network Crossover on NK Landscapes and Spin Glasses

نویسندگان

  • Mark Hauschild
  • Martin Pelikan
چکیده

This paper describes a network crossover operator based on knowledge gathered from either prior problem-specific knowledge or linkage learning methods such as estimation of distribution algorithms (EDAs). This operator can be used in a genetic algorithm (GA) to incorporate linkage in recombination. The performance of GA with network crossover is compared to that of GA with uniform crossover and the hierarchical Bayesian optimization algorithm (hBOA) on 2D Ising spin glasses, NK landscapes, and SK spin glasses. The results are analyzed and discussed.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

On the Eeectiveness of Evolutionary Search in High{dimensional Nk{landscapes

| NK{landscapes ooer the ability to assess the performance of evolutionary algorithms on problems with different degrees of epistasis. In this paper, we study the performance of six algorithms in NK{landscapes with low and high dimension while keeping the amount of epistatic interactions constant. The results show that compared to genetic local search algorithms, the performance of standard gen...

متن کامل

Genetic Algorithms on NK-Landscapes: Effects of Selection, Drift, Mutation, and Recombination

NK-Landscapes are stochastically generated fitness functions on bit strings, parameterized with N bits and K epistatic interactions between bits. The term epistasis describes nonlinearities in fitness functions due to changes in the values of interacting bits. Empirical studies have shown that the overall performance of random bit climbers on NK-Landscapes is superior to the performance of some...

متن کامل

Multi-Parent's Niche: n-ary Crossovers on NK-Landscapes

1 I n t r o d u c t i o n In previous papers the multi-parent scanning crossover and diagonal crossover were tested on numerical optimization problems, [3, 4, 5]. These tests were performed on a test suite of common EA test functions, still they were chosen somewhat arbitrarily, moreover the types of fitness landscapes they span are difficult to characterize. I t was therefore not possible to l...

متن کامل

The Distributional Biases of Crossover Operators

The choice of genetic operators is one way in which genetic algorithms can be tailored to speciic optimization problems. For bit represented problems, the choice of crossover operator , or the choice not to use a crossover operator, can dramatically aaect search performance. The eecacy of crossover for genetic search is governed by the relationship between the crossover biases and the search pr...

متن کامل

Optimal Neuron Selection: NK Echo State Networks for Reinforcement Learning

This paper introduces the NK Echo State Network. The problem of learning in the NK Echo State Network is reduced to the problem of optimizing a special form of a Spin Glass Problem known as an NK Landscape. No weight adjustment is used; all learning is accomplished by spinning up (turning on) or spinning down (turning off) neurons in order to find a combination of neurons that work together to ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2010