CACM editor-in-chief steps down advice for editorial board

In this month’s Communications of the ACM Moshe Vardi, the current editor-in-chief, publishes a one page valedictory article

Comments Off on CACM editor-in-chief steps down advice for editorial board

Parameters, Parameters, Parameters

The practice of evolutionary algorithms involves a mundane yet inescapable phase, namely, finding parameters that work well. How big should the population be? How many generations should the algorithm run? What is the (tournament selection) tournament size? What probabilities should one assign to crossover and mutation? All these nagging questions need good answers if one is to embrace success. Through an extensive series of experiments over multiple evolutionary algorithm implementations and problems we show that parameter space tends to be rife with viable parameters. We aver that this renders the life of the practitioner that much easier, and cap off our study with an advisory digest for the weary.
Wanna learn more? The full paper is here.
Comments Off on Parameters, Parameters, Parameters

CFP: Special Issue on Genetic Programming, Evolutionary Computation and Visualization

Guest Editors: Nadia Boukhelifa and Evelyne Lutton (both at INRA, Versailles-Grignon, France). Please see the full CFP on the Springer site.

Comments Off on CFP: Special Issue on Genetic Programming, Evolutionary Computation and Visualization

IEEE Transactions on Evolutionary Computation Society Information

Comments Off on IEEE Transactions on Evolutionary Computation Society Information

IEEE Transactions on Evolutionary Computation information for authors

Comments Off on IEEE Transactions on Evolutionary Computation information for authors

Investigating the Effect of Imbalance Between Convergence and Diversity in Evolutionary Multiobjective Algorithms

There are two main tasks involved in addressing a multiobjective optimization problem (MOP) by evolutionary multiobjective (EMO) algorithms: 1) make the population converge close to the Pareto-optimal front and 2) maintain adequate population diversity. However, most state-of-the-art EMO algorithms are designed based on the “convergence first and diversity second” principle. It has been observed that although these EMO algorithms have been successful in optimizing many real-world MOPs, they fail to solve certain problems that feature a severe imbalance between diversity preservation and achieving convergence. This paper characterizes an imbalanced MOP by clearly defining properties and indicating the reasons for the existing EMO algorithms’ difficulties in solving them. We then present 14 imbalanced problems, with and without constraints. Computational results using four existing EMO algorithms—elitist non-dominated sorting genetic algorithm (NSGA-II), multiobjective evolutionary algorithm based on decomposition (MOEA/D), strength Pareto evolutionary algorithm 2 (SPEA2), and S metric selection EMO algorithm (SMS-EMOA) and a proposed generalized vector-evaluated genetic algorithm are then presented. It is seen that these EMO algorithms cannot solve these imbalanced problems, but they are able to solve the problems when augmented by multiobjective to multiobjective (M2M), an approach that decomposes the population into several interacting subpopulations. These results and the successful application of the EMO methods with the M2M approach even on standard so-called balanced problems indicate the usefulness of using the M2M approach.

Comments Off on Investigating the Effect of Imbalance Between Convergence and Diversity in Evolutionary Multiobjective Algorithms

IEEE World Congress on Computational Intelligence

Comments Off on IEEE World Congress on Computational Intelligence

Shape Optimization With Surface-Mapped CPPNs

Shape optimization techniques are becoming increasingly important in design and engineering. This growing significance reflects the need to exploit advances in digital fabrication technologies, and the desire to create new types of surface designs for various engineering applications. Evolutionary algorithms (EAs) offer several key advantages for shape optimization, but they can also be restricted, especially as design problems scale up in size. A key challenge for evolutionary shape optimization is to overcome these challenges in order to apply EAs to large-scale, “real-world” engineering problems. This paper presents a new evolutionary approach to shape optimization using what we call “surface-mapped compositional pattern producing networks (CPPNs).” Our method outperforms a state-of-the-art gradient-based method on a simple benchmark problem, and scales well as degrees of freedom are added to the design problem. Our results demonstrate that surface-mapped CPPNs offer practical ways of approaching large-scale, real-world engineering problems with EAs, opening up exciting new opportunities for engineering design.

Comments Off on Shape Optimization With Surface-Mapped CPPNs

How to Pack Trapezoids: Exact and Evolutionary Algorithms

The purposes of this paper are twofold. In the first, we describe an exact polynomial-time algorithm for the pair sequencing problem and show how this method can be used to pack fixed-height trapezoids into a single bin such that interitem wastage is minimized. We then go on to examine how this algorithm can be combined with bespoke evolutionary and local search methods for tackling the multiple-bin version of this problem—one that is closely related to 1-D bin packing. In the course of doing this, a number of ideas surrounding recombination, diversity, and genetic repair are also introduced and analyzed.

Comments Off on How to Pack Trapezoids: Exact and Evolutionary Algorithms

Impacts of Coefficients on Movement Patterns in the Particle Swarm Optimization Algorithm

In this paper, we investigate movement patterns of a particle in the particle swarm optimization (PSO) algorithm. We characterize movement patterns of the particle by two factors: 1) the correlation between its consecutive positions and 2) its range of movement. We introduce the base frequency of movement as a measure for the correlation between positions and the variance of movement as a measure for the range of movement. We determine the base frequency and the variance of movement theoretically and we show how they change with the values of coefficients. We extract a system of equations that enables practitioners to find coefficients’ values to guarantee achieving a given base frequency and variance of movement, i.e., control the movement pattern of particles. We also show that if the base frequency of movement for a particle is small, mid range, or large then the particle’s position at each iteration is positively correlated (smooth movement), uncorrelated (chaotic movement), or negatively correlated (jumping at each iteration) with its previous positions, respectively. We test the effects of the base frequency and variance of movement on the search ability of particles and we show that small base frequencies (i.e., smooth movement) are more effective when the maximum number of function evaluations is large. We found that the most frequently-used coefficient values in PSO literature impose mid-range base frequencies that correspond with a chaotic movement. We also provide new sets of coefficients that outperform existing ones on a set of benchmark functions.

Comments Off on Impacts of Coefficients on Movement Patterns in the Particle Swarm Optimization Algorithm