Search

Novelty search is a powerful tool for finding diverse sets of objects in complicated spaces. Recent experiments on simplified versions of novelty search introduce the idea that novelty search happens at the level of the archive space, rather than individual points. The sparseness measure and archive update criterion create a process that is driven by a two measures: (1) spread out to cover the space while trying to remain as efficiently packed as possible, and (2) metrics inspired by k nearest neighbor theory. In this paper, we generalize previous simplifications of novelty search to include traditional population (μ,λ) dynamics for generating new search points, where the population and the archive are updated separately. We provide some theoretical guidance regarding balancing mutation and sparseness criteria and introduce the concept of saturation as a way of talking about fully covered spaces. We show empirically that claims that novelty search is inherently objectiveless are incorrect. We leverage the understanding of novelty search as an optimizer of archive coverage, suggest several ways to improve the search, and demonstrate one simple improvement-generating some new points directly from the archive rather than the parent population., (© 2023 Massachusetts Institute of Technology.)

Minimizing the number of selected features and maximizing the classification performance are two main objectives in feature selection, which can be formulated as a bi-objective optimization problem. Due to the complex interactions between features, a solution (i.e., feature subset) with poor objective values does not mean that all the features it selects are useless, as some of them combined with other complementary features can greatly improve the classification performance. Thus, it is necessary to consider not only the performance of feature subsets in the objective space, but also their differences in the search space, to explore more promising feature combinations. To this end, this paper proposes a tri-objective method for bi-objective feature selection in classification, which solves a bi-objective feature selection problem as a tri-objective problem by considering the diversity (differences) between feature subsets in the search space as the third objective. The selection based on the converted tri-objective method can maintain a balance between minimizing the number of selected features, maximizing the classification performance, and exploring more promising feature subsets. Furthermore, a novel initialization strategy and an offspring reproduction operator are proposed to promote the diversity of feature subsets in the objective space and improve the search ability, respectively. The proposed algorithm is compared with five multiobjective-based feature selection methods, six typical feature selection methods, and two peer methods with diversity as a helper objective. Experimental results on 20 real-world classification datasets suggest that the proposed method outperforms the compared methods in most scenarios., (© 2023 Massachusetts Institute of Technology.)

In this paper, we compare Bayesian Optimization, Differential Evolution, and an Evolution Strategy employed as a gait-learning algorithm in modular robots. The motivational scenario is the joint evolution of morphologies and controllers, where "newborn" robots also undergo a learning process to optimize their inherited controllers (without changing their bodies). This context raises the question: How do gait-learning algorithms compare when applied to various morphologies that are not known in advance (and thus need to be treated as without priors)? To answer this question, we use a test suite of twenty different robot morphologies to evaluate our gait-learners and compare their efficiency, efficacy, and sensitivity to morphological differences. The results indicate that Bayesian Optimization and Differential Evolution deliver the same solution quality (walking speed for the robot) with fewer evaluations than the Evolution Strategy. Furthermore, the Evolution Strategy is more sensitive for morphological differences (its efficacy varies more between different morphologies) and is more subject to luck (repeated runs on the same morphology show greater variance in the outcomes)., (© 2023 Massachusetts Institute of Technology.)

Exposing an evolutionary algorithm that is used to evolve robot controllers to variable conditions is necessary to obtain solutions which are robust and can cross the reality gap. However, we do not yet have methods for analyzing and understanding the impact of the varying morphological conditions which impact the evolutionary process, and therefore for choosing suitable variation ranges. By morphological conditions, we refer to the starting state of the robot, and to variations in its sensor readings during operation due to noise. In this paper, we introduce a method that permits us to measure the impact of these morphological variations and we analyze the relation between the amplitude of variations, the modality with which they are introduced, and the performance and robustness of evolving agents. Our results demonstrate that (i) the evolutionary algorithm can tolerate morphological variations which have a very high impact, (ii) variations affecting the actions of the agent are tolerated much better than variations affecting the initial state of the agent or of the environment, and (iii) improving the accuracy of the fitness measure through multiple evaluations is not always useful. Moreover, our results show that morphological variations permit generating solutions which perform better both in varying and non-varying conditions., (© 2023 Massachusetts Institute of Technology.)

For offline data-driven multiobjective optimization problems (MOPs), no new data is available during the optimization process. Approximation models (or surrogates) are first built using the provided offline data, and an optimizer, for example, a multiobjective evolutionary algorithm, can then be utilized to find Pareto optimal solutions to the problem with surrogates as objective functions. In contrast to online data-driven MOPs, these surrogates cannot be updated with new data and, hence, the approximation accuracy cannot be improved by considering new data during the optimization process. Gaussian process regression (GPR) models are widely used as surrogates because of their ability to provide uncertainty information. However, building GPRs becomes computationally expensive when the size of the dataset is large. Using sparse GPRs reduces the computational cost of building the surrogates. However, sparse GPRs are not tailored to solve offline data-driven MOPs, where good accuracy of the surrogates is needed near Pareto optimal solutions. Treed GPR (TGPR-MO) surrogates for offline data-driven MOPs with continuous decision variables are proposed in this paper. The proposed surrogates first split the decision space into subregions using regression trees and build GPRs sequentially in regions close to Pareto optimal solutions in the decision space to accurately approximate tradeoffs between the objective functions. TGPR-MO surrogates are computationally inexpensive because GPRs are built only in a smaller region of the decision space utilizing a subset of the data. The TGPR-MO surrogates were tested on distance-based visualizable problems with various data sizes, sampling strategies, numbers of objective functions, and decision variables. Experimental results showed that the TGPR-MO surrogates are computationally cheaper and can handle datasets of large size. Furthermore, TGPR-MO surrogates produced solutions closer to Pareto optimal solutions compared to full GPRs and sparse GPRs., (© 2023 Massachusetts Institute of Technology. Published under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.)

Existing work on offline data-driven optimization mainly focuses on problems in static environments, and little attention has been paid to problems in dynamic environments. Offline data-driven optimization in dynamic environments is a challenging problem because the distribution of collected data varies over time, requiring surrogate models and optimal solutions tracking with time. This paper proposes a knowledge-transfer-based data-driven optimization algorithm to address these issues. First, an ensemble learning method is adopted to train surrogate models to leverage the knowledge of data in historical environments as well as adapt to new environments. Specifically, given data in a new environment, a model is constructed with the new data, and the preserved models of historical environments are further trained with the new data. Then, these models are considered to be base learners and combined as an ensemble surrogate model. After that, all base learners and the ensemble surrogate model are simultaneously optimized in a multitask environment for finding optimal solutions for real fitness functions. In this way, the optimization tasks in the previous environments can be used to accelerate the tracking of the optimum in the current environment. Since the ensemble model is the most accurate surrogate, we assign more individuals to the ensemble surrogate than its base learners. Empirical results on six dynamic optimization benchmark problems demonstrate the effectiveness of the proposed algorithm compared with four state-of-the-art offline data-driven optimization algorithms. Code is available at https://github.com/Peacefulyang/DSE_MFS.git., (© 2023 Massachusetts Institute of Technology.)

This paper reviews recent studies on the Particle Swarm Optimization (PSO) algorithm. The review has been focused on high impact recent articles that have analyzed and/or modified PSO algorithms. This paper also presents some potential areas for future study. [ABSTRACT FROM AUTHOR]

Recently, Rowe and Aishwaryaprajna (2019) introduced a simple majority vote technique that efficiently solves Jump with large gaps, OneMax with large noise, and any monotone function with a polynomial-size image. In this paper, we identify a pathological condition for this algorithm: the presence of spin-flip symmetry in the problem instance. Spin-flip symmetry is the invariance of a pseudo-Boolean function to complementation. Many important combinatorial optimization problems admit objective functions that exhibit this pathology, such as graph problems, Ising models, and variants of propositional satisfiability. We prove that no population size exists that allows the majority vote technique to solve spin-flip symmetric functions of unitation with reasonable probability. To remedy this, we introduce a symmetry-breaking technique that allows the majority vote algorithm to overcome this issue for many landscapes. This technique requires only a minor modification to the original majority vote algorithm to force it to sample strings in {0,1}n from a dimension n-1 hyperplane. We prove a sufficient condition for a spin-flip symmetric function to possess in order for the symmetry-breaking voting algorithm to succeed, and prove its efficiency on generalized TwoMax, a spin-flip symmetric variant of Jump, and families of constructed 3-NAE-SAT and 2-XOR-SAT formulas. We also prove that the algorithm fails on the one-dimensional Ising model, and suggest different techniques for overcoming this. Finally, we present empirical results that explore the tightness of the runtime bounds and the performance of the technique on randomized satisfiability variants., (© 2023 Massachusetts Institute of Technology.)

This paper presents a personal account of the author's 35 years "adventure" with Evolutionary Computation-from the first encounter in 1988 and many years of academic research through to working full-time in business-successfully implementing evolutionary algorithms for some of the world's largest corporations. The paper concludes with some observations and insights., (© 2023 Massachusetts Institute of Technology.)

Thirty years, 1993-2023, is a huge time frame in science. We address some major developments in the field of evolutionary algorithms, with applications in parameter optimization, over these 30 years. These include the covariance matrix adaptation evolution strategy and some fast-growing fields such as multimodal optimization, surrogate-assisted optimization, multiobjective optimization, and automated algorithm design. Moreover, we also discuss particle swarm optimization and differential evolution, which did not exist 30 years ago, either. One of the key arguments made in the paper is that we need fewer algorithms, not more, which, however, is the current trend through continuously claiming paradigms from nature that are suggested to be useful as new optimization algorithms. Moreover, we argue that we need proper benchmarking procedures to sort out whether a newly proposed algorithm is useful or not. We also briefly discuss automated algorithm design approaches, including configurable algorithm design frameworks, as the proposed next step toward designing optimization algorithms automatically, rather than by hand., (© 2023 Massachusetts Institute of Technology.)

NeuroEvolution (NE) refers to a family of methods for optimizing Artificial Neural Networks (ANNs) using Evolutionary Computation (EC) algorithms. NeuroEvolution of Augmenting Topologies (NEAT) is considered one of the most influential algorithms in the field. Eighteen years after its invention, a plethora of methods have been proposed that extend NEAT in different aspects. In this article, we present a systematic literature review (SLR) to list and categorize the methods succeeding NEAT. Our review protocol identified 232 papers by merging the findings of two major electronic databases. Applying criteria that determine the paper's relevance and assess its quality, resulted in 61 methods that are presented in this article. Our review article proposes a new categorization scheme of NEAT's successors into three clusters. NEAT-based methods are categorized based on 1) whether they consider issues specific to the search space or the fitness landscape, 2) whether they combine principles from NE and another domain, or 3) the particular properties of the evolved ANNs. The clustering supports researchers 1) understanding the current state of the art that will enable them, 2) exploring new research directions or 3) benchmarking their proposed method to the state of the art, if they are interested in comparing, and 4) positioning themselves in the domain or 5) selecting a method that is most appropriate for their problem.

This paper addresses the problem of discovering the structure of a fitness function from binary strings to the reals under the assumption of bounded epistasis. Two loci (string positions) are epistatically Linked if the effect of changing the allele (value) at one locus depends on the allele at the other locus. Similarly, a group of loci are epistatically linked if the effect of changing the allele at one locus depends on the alleles at all other loci of the group. Under the assumption that the size of such groups of loci are bounded, and assuming that the function is given only as a "black box function", this paper presents and analyzes a randomized algorithm that finds the complete epistatic structure of the function in the form of the Walsh coefficients of the function. [ABSTRACT FROM AUTHOR]

In the last decade, many works in combinatorial optimisation have shown that, due to the advances in multi-objective optimisation, the algorithms from this field could be used for solving single-objective problems as well. In this sense, a number of papers have proposed multi-objectivising single-objective problems in order to use multi-objective algorithms in their optimisation. In this article, we follow up this idea by presenting a methodology for multi-objectivising combinatorial optimisation problems based on elementary landscape decompositions of their objective function. Under this framework, each of the elementary landscapes obtained from the decomposition is considered as an independent objective function to optimise. In order to illustrate this general methodology, we consider four problems from different domains: the quadratic assignment problem and the linear ordering problem (permutation domain), the 0-1 unconstrained quadratic optimisation problem (binary domain), and the frequency assignment problem (integer domain). We implemented two widely known multi-objective algorithms, NSGA-II and SPEA2, and compared their performance with that of a single-objective GA. The experiments conducted on a large benchmark of instances of the four problems show that the multi-objective algorithms clearly outperform the single-objective approaches. Furthermore, a discussion on the results suggests that the multi-objective space generated by this decomposition enhances the exploration ability, thus permitting NSGA-II and SPEA2 to obtain better results in the majority of the tested instances.

In recent years an increasing number of real-world many-dimensional optimisation problems have been identified across the spectrum of research fields. Many popular evolutionary algorithms use non-dominance as a measure for selecting solutions for future generations. The process of sorting populations into non-dominated fronts is usually the controlling order of computational complexity and can be expensive for large populations or for a high number of objectives. This paper presents two novel-methods for non-dominated sorting: deductive sort and climbing sort. The two new methods are compared to the fast non-dominated sort of NSGA-II and the non-dominated rank sort of the omni-optimizer. The results demonstrate the improved efficiencies of the deductive sort and the reductions in comparisons that can be made when applying inferred dominance relationships defined in this paper. [ABSTRACT FROM AUTHOR]

This paper proposes a (μ + λ)-differential evolution and an improved adaptive trade-off model for solving constrained optimization problems.The proposed (μ + λ)-differential evolution adopts three mutation strategies (i.e., rand/1 strategy, current-to-best/1 strategy, and rand/2 strategy) and binomial crossover to generate the offspring population. Moreover, the current-to-best/1 strategy has been improved in this paper to further enhance the global exploration ability by exploiting the feasibility proportion of the last population. Additionally, the improved adaptive trade-off model includes three main situations: the infeasible situation, the semi-feasible situation, and the feasible situation. In each situation, a constraint-handlingmechanism is designed based on the characteristics of the current population. By combining the (μ + λ)-differential evolution with the improved adaptive trade-off model, a generic method named (μ + λ)-constrained differential evolution ((μ + λ)-CDE) is developed. The (μ + λ)-CDE is utilized to solve 24 well-known benchmark test functions provided for the special session on constrained real-parameter optimization of the 2006 IEEE Congress on Evolutionary Computation (CEC2006). Experimental results suggest that the (μ + λ)-CDE is very promising for constrained optimization, since it can reach the best known solutions for 23 test functions and is able to successfully solve 21 test functions in all runs. Moreover, in this paper, a self-adaptive version of (μ + λ)-CDE is proposed which is the most competitive algorithm so far among the CEC2006 entries. [ABSTRACT FROM AUTHOR]

In evolutionary computation, the fitness function normally measures progress toward an objective in the search space, effectively acting as an objective function. Through deception, such objective functions may actually prevent the objective from being reached. While methods exist to mitigate deception, they leave the underlying pathology untreated: Objective functions themselves may actively misdirect search toward dead ends. This paper proposes an approach to circumventing deception that also yields a new perspective on open-ended evolution. Instead of either explicitly seeking an objective or modeling natural evolution to capture open-endedness, the idea is to simply search for behavioral novelty. Even in an objective-based problem, such novelty search ignores the objective. Because many points in the search space collapse to a single behavior, the search for novelty is often feasible. Furthermore, because there are only so many simple behaviors, the search for novelty leads to increasing complexity. By decoupling open-ended search from artificial life worlds, the search for novelty is applicable to real world problems. Counterintuitively, in the maze navigation and biped walking tasks in this paper, novelty search significantly outperforms objective-based search, suggesting the strange conclusion that some problems are best solved by methods that ignore the objective. The main lesson is the inherent limitation of the objective-based paradigm and the unexploited opportunity to guide search through other means. [ABSTRACT FROM AUTHOR]

This paper proposes a graph-based evolutionary algorithm called Genetic Network Programming (GNP). Our goal is to develop GNP, which can deal with dynamic environments efficiently and effectively, based on the distinguished expression ability of the graph (network) structure. The characteristics of GNP are as follows. 1) GNP programs are composed of a number of nodes which execute simple judgment/ processing, and these nodes are connected by directed links to each other. 2) The graph structure enables GNP to re-use nodes, thus the structure can be very compact. 3) The node transition of GNP is executed according to its node connections without any terminal nodes, thus the past history of the node transition affects the current node to be used and this characteristic works as an implicit memory function. These structural characteristics are useful for dealing with dynamic environments. Furthermore, we propose an extended algorithm, "GNP with Reinforcement Learning (GNPRL)" which combines evolution and reinforcement learning in order to create effective graph structures and obtain better results in dynamic environments. In this paper, we applied GNP to the problem of determining agents' behavior to evaluate its effectiveness. Tileworld was used as the simulation environment. The results show some advantages for GNP over conventional methods. [ABSTRACT FROM AUTHOR]

Steady State models of Evolutionary Algorithms are widely used, yet surprisingly little attention has been paid to the effects arising from different replacement strategies. This paper explores the use of mathematical models to characterise the selection pressures arising in a selection-only environment. The first part brings together models for the behaviour of seven different replacement mechanisms and provides expressions for various proposed indicators of Evolutionary Algorithm behaviour. Some of these have been derived elsewhere, and are included for completeness, but the majority are new to this paper. These theoretical indicators are used to compare the behaviour of the different strategies. The second part of this paper examines the practical relevance of these indicators as predictors for algorithms' relative performance in terms of optimisation time and reliability. It is not the intention of this paper to come up with a "one size fits all" recommendation for choice of replacement strategy. Although some strategies may have little to recommend them, the relative ranking of others is shown to depend on the intended use of the algorithm to be implemented, as reflected in the choice of performance metrics. [ABSTRACT FROM AUTHOR]

Fitness functions derived from certain types of white-box test goals can be inadequate for evolutionary software test data generation (Evolutionary Testing), due to a lack of search guidance to the required test data. Often this is because the fitness function does not take into account data dependencies within the program under test, and the fact that certain program statements may need to have been executed prior to the target structure in order for it to be feasible. This paper proposes a solution to this problem by hybridizing Evolutionary Testing with an extended Chaining Approach. The Chaining Approach is a method which identifies statements on which the target structure is data dependent, and incrementally develops chains of dependencies in an event sequence. By incorporating this facility into Evolutionary Testing, and by performing a test data search for each generated event sequence, the search can be directed into potentially promising, unexplored areas of the test object's input domain. Results presented in the paper show that test data can be found for a number of test goals with this hybrid approach that could not be found by using the original Evolutionary Testing approach alone. One such test goal is drawn from code found in the publicly available libpng library. [ABSTRACT FROM AUTHOR]

The field of Artificial Immune Systems (AIS) concerns the study and development of computationally interesting abstractions of the immune system. This survey tracks the development of AIS since its inception, and then attempts to make an assessment of its usefulness, defined in terms of 'distinctiveness' and 'effectiveness.' In this paper, the standard types of AIS are examined—Negative Selection, Clonal Selection and Immune Networks—as well as a new breed of AIS, based on the immunological 'danger theory.' The paper concludes that all types of AIS largely satisfy the criteria outlined for being useful, but only two types of AIS satisfy both criteria with any certainty. [ABSTRACT FROM AUTHOR]

Memetic algorithms (MAs) have demonstrated very effective in combinatorial optimization. This paper offers explanations as to why this is so by investigating the performance of MAs in terms of efficiency and effectiveness. A special class of MAs is used to discuss efficiency and effectiveness for local search and evolutionary meta-search. It is shown that the efficiency of MAs can be increased drastically with the use of domain knowledge. However, effectiveness highly depends on the structure of the problem. As is well-known, identifying this structure is made easier with the notion of fitness landscapes: the local properties of the fitness landscape strongly influence the effectiveness of the local search while the global properties strongly influence the effectiveness of the evolutionary meta-search. This paper also introduces new techniques for analyzing the fitness landscapes of combinatorial problems; these techniques focus on the investigation of random walks in the fitness landscape starting at locally optimal solutions as well as on the escape from the basins of attractions of current local optima. It is shown for NK-landscapes and landscapes of the unconstrained binary quadratic programming problem (BQP) that a random walk to another local optimum can be used to explain the efficiency of recombination in comparison to mutation. Moreover, the paper shows that other aspects like the size of the basins of attractions of local optima are important for the efficiency of MAs and a local search escape analysis is proposed. These simple analysis techniques have several advantages over previously proposed statistical measures and provide valuable insight into the behaviour of MAs on different kinds of landscapes. [ABSTRACT FROM AUTHOR]

Over the past few years, the research on evolutionary algorithms has demonstrated their niche in solving multiobjective optimization problems, where the goal is to find a number of Pareto-optimal solutions in a single simulation run. Many studies have depicted different ways evolutionary algorithms can progress towards the Pareto-optimal set with a widely spread distribution of solutions. However, none of the multiobjective evolutionary algorithms (MOEAs) has a proof of convergence to the true Pareto-optimal solutions with a wide diversity among the solutions. In this paper, we discuss why a number of earlier MOEAs do not have such properties. Based on the concept of ε-dominance, new archiving strategies are proposed that overcome this fundamental problem and provably lead to MOEAs that have both the desired convergence and distribution properties. A number of modifications to the baseline algorithm are also suggested. The concept of ε-dominance introduced in this paper is practical and should make the proposed algorithms useful to researchers and practitioners alike. [ABSTRACT FROM AUTHOR]

The gene expression process in nature produces different proteins in different cells from different portions of the DNA. Since proteins control almost every important activity in a living organism, at an abstract level, gene expression can be viewed as a process that evaluates the merit or "fitness" of the DNA. This distributed evaluation of the DNA would not be possible without a decomposed representation of the fitness function defined over the DNAs. This paper argues that, unless the living body was provided with such a representation, we have every reason to believe that it must have an efficient mechanism to construct this distributed representation. This paper demonstrates polynomial-time computability of such a representation by proposing a class of efficient algorithms. The main contribution of this paper is two-fold. On the algorithmic side, it offers a way to scale up evolutionary search by detecting the underlying structure of the search space. On the biological side, it proves that the distributed representation of the evolutionary fitness function in gene expression can be computed in polynomial-time. It advances our understanding about the representation construction in gene expression from the perspective of computing. It also presents experimental results supporting the theoretical performance of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Local selection is a simple selection scheme in evolutionary computation. Individual fitnesses are accumulated over time and compared to a fixed threshold, rather than to each other, to decide who gets to reproduce. Local selection, coupled with fitness functions stemming from the consumption of finite shared environmental resources, maintains diversity in a way similar to fitness sharing. However, it is more efficient than fitness sharing and lends itself to parallel implementations for distributed tasks. While local selection is not prone to premature convergence, it applies minimal selection pressure to the population. Local selection is, therefore, particularly suited to Pareto optimization or problem classes where diverse solutions must be covered. This paper introduces ELSA, an evolutionary algorithm employing local selection and outlines three experiments in which ELSA is applied to multiobjective problems: a multimodal graph search problem, and two Pareto optimization problems. In all these experiments, ELSA significantly outperforms other well-known evolutionary algorithms. The paper also discusses scalability, parameter dependence, and the potential distributed applications of the algorithm. [ABSTRACT FROM AUTHOR]

This paper examines the scalability of several types of parallel genetic algorithms (GAs). The objective is to determine the optimal number of processors that can be used by each type to minimize the execution time. The first part of the paper considers algorithms with a single population. The investigation focuses on an implementation where the population is distributed to several processors, but the results are applicable to more common master-slave implementations, where the population is entirely stored in a master processor and multiple slaves are used to evaluate the fitness. The second part of the paper deals with parallel GAs with multiple populations. It first considers a bounding case where the connectivity, the migration rate, and the frequency of migrations are set to their maximal values. Then, arbitrary regular topologies with lower migration rates are considered and the frequency of migrations is set to its lowest value. The investigation is mainly theoretical, but experimental evidence with an additively-decomposable function is included to illustrate the accuracy of the theory. In all cases, the calculations show that the optimal number of processors that minimizes the execution time is directly proportional to the square root of the population size and the fitness evaluation time. Since these two factors usually increase as the domain becomes more difficult, the results of the paper suggest that parallel GAs can integrate large numbers of processors and significantly reduce the execution time of many practical applications. [ABSTRACT FROM AUTHOR]

The hypervolume indicator has frequently been used for comparing evolutionary multi-objective optimization (EMO) algorithms. A reference point is needed for hypervolume calculation. However, its specification has not been discussed in detail from a viewpoint of fair performance comparison. A slightly worse point than the nadir point is usually used for hypervolume calculation in the EMO community. In this paper, we propose a reference point specification method for fair performance comparison of EMO algorithms. First, we discuss the relation between the reference point specification and the optimal distribution of solutions for hypervolume maximization. It is demonstrated that the optimal distribution of solutions strongly depends on the location of the reference point when a multi-objective problem has an inverted triangular Pareto front. Next, we propose a reference point specification method based on theoretical discussions on the optimal distribution of solutions. The basic idea is to specify the reference point so that a set of well-distributed solutions over the entire linear Pareto front has a large hypervolume and all solutions in such a solution set have similar hypervolume contributions. Then, we examine whether the proposed method can appropriately specify the reference point through computational experiments on various test problems. Finally, we examine the usefulness of the proposed method in a hypervolume-based EMO algorithm. Our discussions and experimental results clearly show that a slightly worse point than the nadir point is not always appropriate for performance comparison of EMO algorithms.

This paper presents an effective evolutionary hybrid for solving the permutation flowshop scheduling problem. Based on a memetic algorithm, the procedure uses a construction component that generates initial solutions through the use of a novel reblocking mechanism operating according to a biased random sampling technique. This component is aimed at forcing the operations having smaller processing times to appear on the critical path. The goal of the construction component is to fill an initial pool with high-quality solutions for a memetic algorithm that looks for even higher-quality solutions. In the memetic algorithm, whenever a crossover operator and possibly a mutation are performed, the offspring genome is fine-tuned by a combination of 2-exchange swap and insertion local searches. The same with the employed construction method; in these local searches, the critical path notion has been used to exploit the structure of the problem. The results of computational experiments on the benchmark instances indicate that these components have strong synergy, and their integration has created a robust and effective procedure that outperforms several state-of-the-art procedures on a number of the benchmark instances. By deactivating different components enhancing the evolutionary module of the procedure, the effects of these components have also been examined.

The article discusses various reports published within the issue including one on multi-objective evolutionary optimization, another on the combination of evolutionary algorithms with fuzzy control systems and one on the problem concerning hardware optimizations with evolutionary algorithms.

In this paper we extend a previously proposed randomized landscape generator in combination with a comparative experimental methodology to study the behavior of continuous metaheuristic optimization algorithms. In particular, we generate twodimensional landscapes with parameterized, linear ridge structure, and perform pairwise comparisons of algorithms to gain insight into what kind of problems are easy and difficult for one algorithm instance relative to another.We apply this methodology to investigate the specific issue of explicit dependency modeling in simple continuous estimation of distribution algorithms. Experimental results reveal specific examples of landscapes (with certain identifiable features) where dependency modeling is useful, harmful, or has little impact on mean algorithm performance. Heat maps are used to compare algorithm performance over a large number of landscape instances and algorithm trials. Finally, we perform a meta-search in the landscape parameter space to find landscapes which maximize the performance between algorithms. The results are related to some previous intuition about the behavior of these algorithms, but at the same time lead to new insights into the relationship between dependency modeling in EDAs and the structure of the problem landscape. The landscape generator and overall methodology are quite general and extendable and can be used to examine specific features of other algorithms. [ABSTRACT FROM AUTHOR]

Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that explore the space of potential solutions by building and sampling explicit probabilistic models of promising candidate solutions. While the primary goal of applying EDAs is to discover the global optimum or at least its accurate approximation, besides this, any EDA provides us with a sequence of probabilistic models, which inmost cases hold a great deal of information about the problem. Although using problem-specific knowledge has been shown to significantly improve performance of EDAs and other evolutionary algorithms, this readily available source of problem-specific information has been practically ignored by the EDA community. This paper takes the first step toward the use of probabilistic models obtained by EDAs to speed up the solution of similar problems in the future. More specifically, we propose two approaches to biasing model building in the hierarchical Bayesian optimization algorithm (hBOA) based on knowledge automatically learned from previous hBOA runs on similar problems. We show that the proposed methods lead to substantial speedups and argue that the methods should work well in other applications that require solving a large number of problems with similar structure. [ABSTRACT FROM AUTHOR]

In a multimodal optimization task, the main purpose is to find multiple optimal solutions (global and local), so that the user can have better knowledge about different optimal solutions in the search space and as and when needed, the current solution may be switched to another suitable optimum solution. To this end, evolutionary optimization algorithms (EA) stand as viable methodologies mainly due to their ability to find and capture multiple solutions within a population in a single simulation run. With the preselection method suggested in 1970, there has been a steady suggestion of new algorithms. Most of these methodologies employed a niching scheme in an existing single-objective evolutionary algorithm framework so that similar solutions in a population are de-emphasized in order to focus and maintain multiple distant yet near-optimal solutions. In this paper, we use a completely different strategy in which the single-objective multimodal optimization problem is converted into a suitable bi-objective optimization problem so that all optimal solutions become members of the resulting weak Pareto-optimal set. With the modified definitions of domination and different formulations of an artificially created additional objective function, we present successful results on problems with as large as 500 optima. Most past multimodal EA studies considered problems having only a few variables. In this paper, we have solved up to 16-variable test problems having as many as 48 optimal solutions and for the first time suggested multimodal constrained test problems which are scalable in terms of number of optima, constraints, and variables. The concept of using bi-objective optimization for solving single-objective multimodal optimization problems seems novel and interesting, and more importantly opens up further avenues for research and application. [ABSTRACT FROM AUTHOR]

In a multimodal optimization task, the main purpose is to find multiple optimal solutions (global and local), so that the user can have a better knowledge about different optimal solutions in the search space and as and when needed, the current solution may be switched to another suitable optimum solution. To this end, evolutionary optimization algorithms (EA) stand as viable methodologies mainly due to their ability to find and capture multiple solutions within a population in a single simulation run. With the preselection method suggested in 1970, there has been a steady suggestion of new algorithms. Most of these methodologies employed a niching scheme in an existing single-objective evolutionary algorithm framework so that similar solutions in a population are de-emphasized in order to focus and maintain multiple distant yet near-optimal solutions. In this paper, we use a completely different strategy in which the single-objective multimodal optimization problem is converted into a suitable bi-objective optimization problem so that all optimal solutions become members of the resulting weak Pareto-optimal set. With the modified definitions of domination and different formulations of an artificially created additional objective function, we present successful results on problems with as large as 500-optima. Most past multimodal EA studies considered problems having a few variables. In this paper, we have solved up to 16-variable test-problems having as many as 48 optimal solutions and for the first time suggested multimodal constrained test-problems which are scalable in terms of number of optima, constraints, and variables. The concept of using bi-objective optimization for solving single-objective multimodal optimization problems seems novel and interesting, and more importantly opens up further avenues for research and application. [ABSTRACT FROM AUTHOR]

In this paper we obtain bounds on the probability of convergence to the optimal solution for the compact genetic algorithm (cGA) and the population based incremental learning (PBIL). Moreover, we give a sufficient condition for convergence of these algorithms to the optimal solution and compute a range of possible values for algorithm parameters at which there is convergence to the optimal solution with a predefined confidence level. [ABSTRACT FROM AUTHOR]

Mutation applied indiscriminately across a population has, on average, a detrimental effect on the accumulation of solution alleles within the population and is usually beneficial only when targeted at individuals with few solution alleles. Many common selection techniques can delete individuals with more solution alleles than are easily recovered by mutation. The paper identifies static and dynamic selection thresholds governing accumulation of information in a genetic algorithm (GA).When individuals are ranked by fitness, there exists a dynamic threshold defined by the solution density of surviving individuals and a lower static threshold defined by the solution density of the information source used for mutation. Replacing individuals ranked below the static threshold with randomly generated individuals avoids the need for mutation while maintaining diversity in the population with a consequent improvement in population fitness. By replacing individuals ranked between the thresholds with randomly selected individuals from above the dynamic threshold, population fitness improves dramatically. We model the dynamic behavior of GAs using these thresholds and demonstrate their effectiveness by simulation and benchmark problems. [ABSTRACT FROM AUTHOR]

Memetic algorithms with continuous local search methods have arisen as effective tools to address the difficulty of obtaining reliable solutions of high precision for complex continuous optimisation problems. There exists a group of continuous local search algorithms that stand out as exceptional local search optimisers. However, on some occasions, they may become very expensive, because of the way they exploit local information to guide the search process. In this paper, they are called intensive continuous local search methods. Given the potential of this type of local optimisation methods, it is interesting to build prospective memetic algorithm models with them. This paper presents the concept of local search chain as a springboard to design memetic algorithm approaches that can effectively use intense continuous local search methods as local search operators. Local search chain concerns the idea that, at one stage, the local search operator may continue the operation of a previous invocation, starting from the final configuration (initial solution, strategy parameter values, internal variables, etc.) reached by this one. The proposed memetic algorithm favours the formation of local search chains during the memetic algorithm run with the aim of concentrating local tuning in search regions showing promise. In order to study the performance of the newmemetic algorithm model, an instance is implemented with CMA-ES as an intense local search method. The benefits of the proposal in comparison to other kinds of memetic algorithms and evolutionary algorithms proposed in the literature to deal with continuous optimisation problems are experimentally shown. Concretely, the empirical study reveals a clear superiority when tackling high-dimensional problems. [ABSTRACT FROM AUTHOR]

In this paper, we discuss the idea of incorporating preference information into evolutionary multi-objective optimization and propose a preference-based evolutionary approach that can be used as an integral part of an interactive algorithm. One algorithm is proposed in the paper. At each iteration, the decision maker is asked to give preference information in terms of his or her reference point consisting of desirable aspiration levels for objective functions. The information is used in an evolutionary algorithm to generate a new population by combining the fitness function and an achievement scalarizing function. In multi-objective optimization, achievement scalarizing functions are widely used to project a given reference point into the Pareto optimal set. In our approach, the next population is thus more concentrated in the area where more preferred alternatives are assumed to lie and the whole Pareto optimal set does not have to be generated with equal accuracy. The approach is demonstrated by numerical examples. [ABSTRACT FROM AUTHOR]

The evolutionary algorithm stochastic process is well-known to be Markovian. These have been under investigation in much of the theoretical evolutionary computing research. When the mutation rate is positive, the Markov chain modeling of an evolutionary algorithm is irreducible and, therefore, has a unique stationary distribution. Rather little is known about the stationary distribution. In fact, the only quantitative facts established so far tell us that the stationary distributions of Markov chains modeling evolutionary algorithms concentrate on uniform populations (i.e., those populations consisting of a repeated copy of the same individual). At the same time, knowing the stationary distribution may provide some information about the expected time it takes for the algorithm to reach a certain solution, assessment of the biases due to recombination and selection, and is of importance in population genetics to assess what is called a "genetic load" (see the introduction for more details). In the recent joint works of the first author, some bounds have been established on the rates at which the stationary distribution concentrates on the uniform populations. The primary tool used in these papers is the "quotient construction" method. It turns out that the quotient construction method can be exploited to derive much more informative bounds on ratios of the stationary distribution values of various subsets of the state space. In fact, some of the bounds obtained in the current work are expressed in terms of the parameters involved in all the three main stages of an evolutionary algorithm: namely, selection, recombination, and mutation. [ABSTRACT FROM AUTHOR]

A continuous location problem in which a firm wants to set up two or more new facilities in a competitive environment is considered. Other facilities offering the same product or service already exist in the area. Both the locations and the qualities of the new facilities are to be found so as to maximize the profit obtained by the firm. This is a global optimization problem, with many local optima. In this paper we analyze several approaches to solve it, namely, three multistart local search heuristics, a multistart simulated annealing algorithm, and two variants of an evolutionary algorithm. Through a comprehensive computational study it is shown that the evolutionary algorithms are the heuristics that provide the best solutions. Furthermore, using a set of problems for which the optimal solutions are known, only the evolutionary algorithms were able to find the optimal solutions for all the instances. The evolutionary strategies presented in this paper can be easily adapted to handle other continuous location problems. [ABSTRACT FROM AUTHOR]

This paper proposes a local search optimizer that, employed as an additional operator in multiobjective evolutionary techniques, can help to find more precise estimates of the Pareto-optimal surface with a smaller cost of function evaluation. The new operator employs quadratic approximations of the objective functions and constraints, which are built using only the function samples already produced by the usual evolutionary algorithm function evaluations. The local search phase consists of solving the auxiliary multiobjective quadratic optimization problem defined from the quadratic approximations, scalarized via a goal attainment formulation using an LMI solver. As the determination of the new approximated solutions is performed without the need of any additional function evaluation, the proposed methodology is suitable for costly black-box optimization problems. [ABSTRACT FROM AUTHOR]

Various methods have been defined to measure the hardness of a fitness function for evolutionary algorithms and other black-box heuristics. Examples include fitness landscape analysis, epistasis, fitness-distance correlations etc., all of which are relatively easy to describe. However, they do not always correctly specify the hardness of the function. Some measures are easy to implement, others are more intuitive and hard to formalize. This paper rigorously defines difficulty measures in black-box optimization and proposes a classification. Different types of realizations of such measures are studied, namely exact and approximate ones. For both types of realizations, it is proven that predictive versions that run in polynomial time in general do not exist unless certain complexity-theoretical assumptions are wrong. [ABSTRACT FROM AUTHOR]

It has been empirically established that multiobjective evolutionary algorithms do not scale well with the number of conflicting objectives. This paper shows that the convergence rate of all comparison-based multi-objective algorithms, for the Hausdorff distance, is not much better than the convergence rate of the random search under certain conditions. The number of objectives must be very moderate and the framework should hold the following assumptions: the objectives are conflicting and the computational cost is lower bounded by the number of comparisons is a good model. Our conclusions are: (i) the number of conflicting objectives is relevant (ii) the criteria based on comparisons with random-search for multi-objective optimization is also relevant (iii) having more than 3-objectives optimization is very hard. Furthermore, we provide some insight into cross-over operators. [ABSTRACT FROM AUTHOR]

This paper demonstrates how evolutionary computation can be used to acquire difficult to solve combinatorial problem instances. As a result of this technique, the corresponding algorithms used to solve these instances are stress-tested. The technique is applied in three important domains of combinatorial optimisation, binary constraint satisfaction, Boolean satisfiability, and the travelling salesman problem. The problem instances acquired through this technique are more difficult than the ones found in popular benchmarks. In this paper, these evolved instances are analysed with the aim to explain their difficulty in terms of structural properties, thereby exposing the weaknesses of corresponding algorithms. [ABSTRACT FROM AUTHOR]

This paper investigates σ-self-adaptation for real valued evolutionary algorithms on linear fitness functions. We identify the step-size logarithm log σ as a key quantity to understand strategy behavior. Knowing the bias of mutation, recombination, and selection on log σ is sufficient to explain σ-dynamics and strategy behavior in many cases, even from previously reported results on non-linear and/or noisy fitness functions. On a linear fitness function, if intermediate multi-recombination is applied on the object parameters, the i-th best and the i-th worst individual have the same σ- distribution. Consequently, the correlation between fitness and step-size σ is zero. Assuming additionally that σ-changes due to mutation and recombination are unbiased, then σ-self-adaptation enlarges σ if and only if μ < λ/2, given (μ, λ)-truncation selection. Experiments show the relevance of the given assumptions. [ABSTRACT FROM AUTHOR]

Evolutionary algorithms are frequently applied to dynamic optimization problems in which the objective varies with time. It is desirable to gain an improved understanding of the influence of different genetic operators and of the parameters of a strategy on its tracking performance. An approach that has proven useful in the past is to mathematically analyze the strategy's behavior in simple, idealized environments. The present paper investigates the performance of a multiparent evolution strategy that employs cumulative step length adaptation for an optimization task in which the target moves linearly with uniform speed. Scaling laws that quite accurately describe the behavior of the strategy and that greatly contribute to its understanding are derived. It is shown that in contrast to previously obtained results for a randomly moving target, cumulative step length adaptation fails to achieve optimal step lengths if the target moves in a linear fashion. Implications for the choice of population size parameters are discussed. [ABSTRACT FROM AUTHOR]

Although Evolutionary Algorithms (EAs) have been successfully applied to optimization in discrete search spaces, theoretical developments remain weak, in particular for population-based EAs. This paper presents a first rigorous analysis of the (μ+1) EA on pseudo-Boolean functions. Using three well-known example functions from the analysis of the (1+1) EA, we derive bounds on the expected runtime and success probability. For two of these functions, upper and lower bounds on the expected runtime are tight, and on all three functions, the (μ+1) EA is never more efficient than the (1+1) EA. Moreover, all lower bounds grow with μ. On a more complicated function, however, a small increase of μ provably decreases the expected runtime drastically. This paper develops a new proof technique that bounds the runtime of the (μ+1) EA. It investigates the stochastic process for creating family trees of individuals; the depth of these trees is bounded. Thereby, the progress of the population towards the optimum is captured. This new technique is general enough to be applied to other population-based EAs. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the space complexity of the Estimation of Distribution Algorithms (EDAs), a class of sampling-based variants of the genetic algorithm. By analyzing the nature of EDAs, we identify criteria that characterize the space complexity of two typical implementation schemes of EDAs, the factorized distribution algorithm and Bayesian network-based algorithms. Using random additive functions as the prototype, we prove that the space complexity of the factorized distribution algorithm and Bayesian network-based algorithms is exponential in the problem size even if the optimization problem has a very sparse interaction structure. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the use of a self-adaptive Pareto evolutionary multi-objective optimization (EMO) approach for evolving the controllers of virtual embodied organisms. The objective of this paper is to demonstrate the trade-off between quality of solutions and computational cost. We show empirically that evolving controllers using the proposed algorithm incurs significantly less computational cost when compared to a self-adaptive weighted sum EMO algorithm, a self-adaptive single-objective evolutionary algorithm (EA) and a hand-tuned Pareto EMO algorithm. The main contribution of the self-adaptive Pareto EMO approach is its ability to produce sufficiently good controllers with different locomotion capabilities in a single run, thereby reducing the evolutionary computational cost and allowing the designer to explore the space of good solutions simultaneously. Our results also show that self-adaptation was found to be highly beneficial in reducing redundancy when compared against the other algorithms. Moreover, it was also shown that genetic diversity was being maintained naturally by virtue of the system's inherent multi-objectivity. [ABSTRACT FROM AUTHOR]

The evolutionary learning mechanism in XCS strongly depends on its accuracy-based fitness approach. The approach is meant to result in an evolutionary drive from classifiers of low accuracy to those of high accuracy. Since, given inaccuracy, lower specificity often corresponds to lower accuracy, fitness pressure most often also results in a pressure towards higher specificity. Moreover, fitness pressure should cause the evolutionary process to be innovative in that it combines low-order building blocks of lower accurate classifiers, to higher-order building blocks with higher accuracy. This paper investigates how, when, and where accuracy-based fitness results in successful rule evolution in XCS. Along the way, a weakness in the current proportionate selection method in XCS is identified. Several problem bounds are derived that need to be obeyed to enable proper evolutionary pressure. Moreover, a fitness dilemma is identified that causes accuracy-based fitness to be misleading. Improvements are introduced to XCS to make fitness pressure more robust and overcome the fitness dilemma. Specifically, (1) tournament selection results in a much better fitness-bias exploitation, and (2) bilateral accuracy prevents the fitness dilemma. While the improvements stand for themselves, we believe they also contribute to the ultimate goal of an evolutionary learning system that is able to solve decomposable machine-learning problems quickly, accurately,and reliably. The paper also contributes to the further understanding of XCS in general and the fitness approach in XCS in particular. [ABSTRACT FROM AUTHOR]

This paper analyzes the impact of using noisy data sets in Pittsburgh-style learning classifier systems. This study was done using a particular kind of learning classifier system based on multiobjective selection. Our goal was to characterize the behavior of this kind of algorithms when dealing with noisy domains. For this reason, we developed a theoretical model for predicting the minimal achievable error in noisy domains. Combining this theoretical model for crisp learners with graphical representations of the evolved hypotheses through multiobjective techniques, we are able to bound the behavior of a learning classifier system. This kind of modeling lets us identify relevant characteristics of the evolved hypotheses, such as overfitting conditions that lead to hypotheses that poorly generalize the concept to be learned. [ABSTRACT FROM AUTHOR]

Recent convergence analyses of evolutionary pattern search algorithms (EPSAs) have shown that these methods have a weak stationary point convergence theory for a broad class of unconstrained and linearly constrained problems. This paper describes how the convergence theory for EPSAs can be adapted to allow each individual in a population to have its own mutation step length (similar to the design of evolutionary programing and evolution strategies algorithms). These are called locally-adaptive EPSAs (LA-EPSAs) since each individual's mutation step length is independently adapted in different local neighborhoods. The paper also describes a variety of standard formulations of evolutionary algorithms that can be used for LA-EPSAs. Further, it is shown how this convergence theory can be applied to memetic EPSAs, which use local search to refine points within each iteration. [ABSTRACT FROM AUTHOR]