Faster black-box algorithms through higher arity operators

  title={Faster black-box algorithms through higher arity operators},
  author={Benjamin Doerr and Daniel Johannsen and Timo K{\"o}tzing and P. Lehre and Markus Wagner and Carola Doerr},
  booktitle={Foundations of Genetic Algorithms},
We extend the work of Lehre and Witt (GECCO 2010) on the unbiased black-box model by considering higher arity variation operators. In particular, we show that already for binary operators the black-box complexity of LeadingOnes drops from Θ(<i>n</i><sup>2</sup>) for unary operators to <i>O</i>(<i>n</i> log <i>n</i>). For OneMax, the Ω(<i>n</i> log <i>n</i>) unary black-box complexity drops to <i>O</i>(<i>n</i>) in the binary case. For <i>k</i>-ary operators, <i>k</i> ≤ <i>n</i>, the OneMax… 

Tables from this paper

Reducing the arity in unbiased black-box complexity

It is shown that the power of higher arity operators is much stronger than what the previous O(n/k) bound by Doerr et al. suggests, and the key to this result is an encoding strategy, which might be of independent interest.

Better fixed-arity unbiased black-box algorithms

This paper proposes an alternative strategy for achieving this unbiased black-box complexity of OneMax when 3 ≤ k ≤ log2 n, and constitutes an algorithm which is conceptually simpler than the one by Doerr and Doerr, and in the same time achieves better constant multiples in the asymptotic notation.

The (1+1) Elitist Black-Box Complexity of LeadingOnes

The permutation- and bit-invariant version of LeadingOnes is regarded and it is proved that its (1+1) elitist black-box complexity is Ω(n2), a bound that is matched by (1-1)-type evolutionary algorithms.

Parallel Black-Box Complexity With Tail Bounds

The main result is a general performance limit: it is proved that on every function, the typical optimization time on unimodal and multimodal problems, for the time to find any local optimum, and for the times to even get close to any optimum are reduced.

OneMax in Black-Box Models with Several Restrictions

This work shows that the (1+1) memory-restricted ranking-based black-box complexity of OneMax is linear, and provides improved lower bounds for the complexity of the OneMax in the regarded models.

The $$(1+1)$$(1+1) Elitist Black-Box Complexity of LeadingOnes

The permutation- and bit-invariant version of LeadingOnes is regarded and it is proved that its(1+1) elitist black-box complexity is VarOmega (n^2)Ω(n2), a bound that is matched by(1-1)-type evolutionary algorithms, a bound which shows that for LeadingOns the memory-restriction, together with the selection requirement, has a substantial impact on the best possible performance.

Black-Box Complexity: Breaking the O(n logn) Barrier of LeadingOnes

We show that the unrestricted black-box complexity of the n-dimensional XOR- and permutation-invariant LeadingOnes function class is O(n log(n) / loglogn). This shows that the recent natural looking

Unbiased black-box complexities of jump functions: how to cross large plateaus

It is shown that when the jump size is (1/2 - epsilon)n, that is, only a small constant fraction of the fitness values is visible, then the unbiased black-box complexities for arities 3 and higher are of the same order as those for the simple OneMax function.

Too fast unbiased black-box algorithms

This work gives mutation-only unbiased black-box algorithms having complexity O(n log n) for the classical JumpK test function class and for a subclass of the well-known Partition problem.



Black-Box Search by Unbiased Variation

This paper introduces a more restricted black-box model for optimisation of pseudo-Boolean functions which it is claimed captures the working principles of many randomised search heuristics including simulated annealing, evolutionary algorithms, randomised local search, and others.

Upper and Lower Bounds for Randomized Search Heuristics in Black-Box Optimization

Lower bounds on the black-box complexity of problems are derived without complexity theoretical assumptions and are compared with upper bounds in this scenario.

Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation

This book presents an introduction to Evolutionary Algorithms, a meta-language for programming with real-time implications, and some examples of how different types of algorithms can be tuned for different levels of integration.

On the analysis of the (1+1) evolutionary algorithm

Crossover can provably be useful in evolutionary computation

This is the first theoretical analysis proving the usefulness of crossover for a non-artificial problem and it is shown that the natural evolutionary algorithm for the all-pairs shortest path problem is significantly faster with a crossover operator than without.

Swarm Intelligence

An iterative particle swarm embedding algorithm (PSEA) that learns embeddings of low-dimensional representations for high-dimensional input patterns and achieves competitive or even better embedding like the related methods locally linear embedding, and ISOMAP.

Ant colony optimization

A forage harvester includes four multiple-bladed rotary cutting segments positioned near a shearbar, each including a permanent magnet and a sensing coil to monitor only selected one or ones of the cutting segments.

Black-box search by elimination of fitness functions

Though in its early stages, it is believed that there is utility in search methods based on ideas from the elimination of functions method, and that the viewpoint provides promise and new insight about black-box optimization.

Introduction to algorithms [2nd ed.]