Affine OneMax

  title={Affine OneMax},
  author={A. Berny},
  journal={Proceedings of the Genetic and Evolutionary Computation Conference Companion},
  • A. Berny
  • Published 2021
  • Computer Science, Mathematics
  • Proceedings of the Genetic and Evolutionary Computation Conference Companion
A new class of test functions for black box optimization is introduced. Affine OneMax (AOM) functions are defined as compositions of OneMax and invertible affine maps on bit vectors. The black box complexity of the class is upper bounded by a polynomial of large degree in the dimension. Tunable complexity is achieved by expressing invertible linear maps as finite products of transvections. Finally, experimental results are given to illustrate the performance of search algorithms on AOM… Expand

Figures from this paper


Representational issues in genetic optimization
Formal functions are partially characterized as easy or hard for genetic algorithms to optimize, and affine transformations are shown to be sufficiently powerful to transform at least selected deceptive problems into easy ones. Expand
Polynomials, Basis Sets, and Deceptiveness in Genetic Algorithms
Three function classes are characterized that are fully easy for the genetic algorithm in terms of the polynomial representation, and the relationship between these representations and the more standard Walsh polynomia ls is shown. Expand
Polynomial Time Summary Statistics for a Generalization of MAXSAT
This paper exploits the similar structure of the two MAXSAT problems to create an encompassing class of problems called embedded landscapes, and uses Walsh analysis to explore the nonlinear bit interactions of these important test functions. Expand
Learning decision trees using the Fourier spectrum
The authors demonstrate that any functionf whose L -norm is polynomial can be approximated by a polynomially sparse function, and prove that boolean decision trees with linear operations are a subset of this class of functions. Expand
Almost Tight Upper Bound for Finding Fourier Coefficients of Bounded Pseudo- Boolean Functions
This work considers the problem of finding the Fourier coefficients of k-bounded functions, or equivalently, finding the coefficients of multilinear polynomials on {-1,1}^n of degree k or less, and gives an optimal algorithm for the problem. Expand
The No Free Lunch and problem description length
The No Free Lunch theorem is reviewed and cast within a simple framework for black-box search. A duality result which relates functions being optimized to algorithms optimizing them is obtained andExpand
On the application of linear transformations for genetic algorithms optimization
New results about the effect of linear transformations over the representation, the spectrum and the performance of standard genetic algorithms (GAs) are discussed and an heuristic for construction of non-unitary LTs is proposed. Expand
Test Function Generators as Embedded Landscapes
A parameterized model for the construction of test function generators, called embedded landscapes, is introduced that can be used to isolate the features of combinatorial optimization problems for more control during experimentation and shows that common forms of embedded landscapes allow for a polynomial time Walsh analysis. Expand
Complexity Theory for Discrete Black-Box Optimization Heuristics
Running-time analysis is aimed at understanding the performance of a given heuristic on a given problem by bounding the number of function evaluations that are needed by the heuristic to identify a solution of a desired quality. Expand
Fast and Efficient Black Box Optimization Using the Parameter-less Population Pyramid
It is concluded that P3 is an efficient, general, parameter-less approach to black box optimization which is more effective than existing state-of-the-art techniques. Expand