scientific article; zbMATH DE number 12464
The paper investigates the application of Walsh functions to the analysis of genetic algorithms and introduces Bethke's Walsh-schema transform through the Walsh polynomials. The introduction of a polynomial form presents a common language for discussing the usual composition of mappings that occurs in genetic algorithms when coding are decoded, mapped, and passed as parameters to some objective function. This form provides an intuitive basis for visualizing the interactions of functions and codings using straightforward algebraic methods. The use of polynomials makes proofs and algebraic manipulation more direct and intuitive. The Walsh functions are presented in monomial form and fitness functions are written as a weighted sum of the Walsh functions. The analysis techniques are illustrated by several specific examples, which have demonstrated the usefulness of the methods in determining whether particular coding-function combinations should be easy or hard for a genetic algorithm. A method has been used to design fully deceptive problems. Pascal code implementing the fast Walsh transform is given as an appendix to the paper.
- Fourier and Taylor series on fitness landscapes
- An analysis of a reordering operator on a GA-hard problem
- Hadamard design and artificial neural nets
- System optimization of turbofan engines using genetic algorithms
- Using a genetic algorithm to determine an index of leading economic indicators
- Construction of high-order deceptive functions using low-order Walsh coefficients
- Deception, dominance and implicit parallelism in genetic search.
- Tracking primary hyperplane competitors during genetic search
- GA performance distributions and randomly generated binary constraint satisfaction problems.
- Sufficient conditions for deceptive and easy binary functions
- A proof of the Vose-Liepins conjecture
- Fourier analysis of genetic algorithms
- Macroscopic kinetic equation for a genetic algorithm
- The allocation of storage space for transshipment in vehicle distribution
- Coupled map lattices as spatio-temporal fitness functions: Landscape measures and evolutionary optimization
- An alternative approach for neural network evolution with a genetic algorithm: crossover by combinatorial optimization
- New approaches for the nesting of two-dimensional shapes for press tool design
- Computing epistasis of template functions through Walsh transforms
- Studying the Effects of Dual Coding on the Adaptation of Representation for Linkage in Evolutionary Algorithms
- Evolutionary optimization: pitfalls and booby traps
- scientific article; zbMATH DE number 12465 (Why is no real title available?)
- scientific article; zbMATH DE number 21233 (Why is no real title available?)
- scientific article; zbMATH DE number 710881 (Why is no real title available?)
- Faster learning of control parameters through sharing experiences of autonomous mobile robots
- The Markov network fitness model
- ADAPTATION CHANNELS
- On dynamic policies for quality, price and advertising
- Transit sets of two-point crossover
- Transit sets of k-point crossover operators
- Multi-output parameter estimation of dynamic systems by output shapes
- The 2-epistasis of fitness functions
- A striking property of genetic code-like transformations.
- Novel 2D TOMOCOMD-CARDD molecular descriptors: atom-based stochastic and non-stochastic bilinear indices and their QSPR applications
- Optimisation via encodings: a renormalisation group perspective
- Genetic set recombination and its application to neural network topology optimisation.
- The dynamics of a genetic algorithm for simple random Ising systems
- Hadamard matrices of genetic code and trigonometric functions
- A randomized heuristic repair for the multidimensional knapsack problem
- Vectorial tolerance allocation of bevel gear by discrete optimization
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3978554)