GPEM 16(1) available

As Bill notes below, GPEM 16(1) is now available. It contains:

“Editorial Introduction”
by Lee Spector

“Training genetic programming classifiers by vicinal-risk minimization”
by Ji Ni & Peter Rockett

“Improving GP generalization: a variance-based layered learning approach”
by Maryam Amir Haeri, Mohammad Mehdi Ebadzadeh & Gianluigi Folino

“GA-based approach to find the stabilizers of a given sub-space”
by Mahboobeh Houshmand, Morteza Saheb Zamani, Mehdi Sedighi & Monireh Houshmand

Letter
“A C++ framework for geometric semantic genetic programming”
by Mauro Castelli, Sara Silva & Leonardo Vanneschi

Letter
“Introducing a cross platform open source Cartesian Genetic Programming library”
by Andrew James Turner & Julian Francis Miller

“Acknowledgment”
by L. Spector

Comments Off

GPEM 16(1) Added to GP bibliography

The first issue of the 2015 volume is now available on the springer web pages
and incorporated into the genetic programming bibliography.
Bill

Comments Off

IEEE Transactions on Evolutionary Computation information for authors

Comments Off

Learning Value Functions in Interactive Evolutionary Multiobjective Optimization

This paper proposes an interactive multiobjective evolutionary algorithm (MOEA) that attempts to learn a value function capturing the users’ true preferences. At regular intervals, the user is asked to rank a single pair of solutions. This information is used to update the algorithm’s internal value function model, and the model is used in subsequent generations to rank solutions incomparable according to dominance. This speeds up evolution toward the region of the Pareto front that is most desirable to the user. We take into account the most general additive value function as a preference model and we empirically compare different ways to identify the value function that seems to be the most representative with respect to the given preference information, different types of user preferences, and different ways to use the learned value function in the MOEA. Results on a number of different scenarios suggest that the proposed algorithm works well over a range of benchmark problems and types of user preferences.

Comments Off

IEEE World Congress on Computational Intelligence

Comments Off

Exploratory Landscape Analysis of Continuous Space Optimization Problems Using Information Content

Data-driven analysis methods, such as the information content of a fitness sequence, characterize a discrete fitness landscape by quantifying its smoothness, ruggedness, or neutrality. However, enhancements to the information content method are required when dealing with continuous fitness landscapes. One typically employed adaptation is to sample the fitness landscape using random walks with variable step size. However, this adaptation has significant limitations: random walks may produce biased samples, and uncertainty is added because the distance between observations is not accounted for. In this paper, we introduce a robust information content-based method for continuous fitness landscapes, which addresses these limitations. Our method generates four measures related to the landscape features. Numerical simulations are used to evaluate the efficacy of the proposed method. We calculate the Pearson correlation coefficient between the new measures and other well-known exploratory landscape analysis measures. Significant differences on the measures between benchmark functions are subsequently identified. We then demonstrate the practical relevance of the new measures using them as class predictors on a machine learning model, which classifies the benchmark functions into five groups. Classification accuracy greater than 90% was obtained, with computational costs bounded between 1% and 10% of the maximum function evaluation budget. The results demonstrate that our method provides relevant information, at a low cost in terms of function evaluations.

Comments Off

History-Based Topological Speciation for Multimodal Optimization

Evolutionary algorithms integrating various niching techniques have been widely used to find multiple optima of an optimization problem. In recent years, an increasing amount of research has been focused on the design and application of speciation-based niching techniques. These techniques rely on speciation to partition a population into subpopulations (species) such that each occupies a different region of attraction (niche) on the fitness landscape. Existing speciation methods are either distance-based or topology-based. Topology-based methods are more flexible and have fewer assumptions than distance-based methods. However, existing topology-based methods all require sampling and evaluating new individuals in order to capture the landscape topography. This incurs additional fitness evaluations (FEs), which is a drawback, especially when the FE budget is limited. In this paper, a new topology-based speciation method named history-based topological speciation (HTS) is proposed. It relies exclusively on search history to capture the landscape topography and, therefore, does not require any additional FEs to be performed. To the best of our knowledge, HTS is the only parameter-free speciation method at the moment. Both theoretical and empirical analyses have been conducted. Theoretical analysis shows that HTS incurs acceptable computational overhead. In the experimental study, HTS outperformed existing topology-based methods on benchmark functions in up to 32-D space and with as many as 50 optima, and the time overhead was practically negligible if a single FE took seconds.

Comments Off

A New Local Search-Based Multiobjective Optimization Algorithm

In this paper, a new multiobjective optimization framework based on nondominated sorting and local search (NSLS) is introduced. The NSLS is based on iterations. At each iteration, given a population ${P}$ , a simple local search method is used to get a better population $P{‘}$ , and then the nondominated sorting is adopted on $P cup P{‘}$ to obtain a new population for the next iteration. Furthermore, the farthest-candidate approach is combined with the nondominated sorting to choose the new population for improving the diversity. Additionally, another version of NSLS (NSLS-C) is used for comparison, which replaces the farthest-candidate method with the crowded comparison mechanism presented in the nondominated sorting genetic algorithm II (NSGA-II). The proposed method (NSLS) is compared with NSLS-C and the other three classic algorithms: NSGA-II, MOEA/D-DE, and MODEA on a set of seventeen bi-objective and three tri-objective test problems. The experimental results indicate that the proposed NSLS is able to find a better spread of solutions and a better convergence to the true Pareto-optimal front compared to the other four algorithms. Furthermore, the sensitivity of NSLS is also experimentally investigated in this paper.

Comments Off

The Rolling Tide Evolutionary Algorithm: A Multiobjective Optimizer for Noisy Optimization Problems

As the methods for evolutionary multiobjective optimization (EMO) mature and are applied to a greater number of real-world problems, there has been gathering interest in the effect of uncertainty and noise on multiobjective optimization, specifically how algorithms are affected by it, how to mitigate its effects, and whether some optimizers are better suited to dealing with it than others. Here we address the problem of uncertain evaluation, in which the uncertainty can be modeled as an additive noise in objective space. We develop a novel algorithm, the rolling tide evolutionary algorithm (RTEA), which progressively improves the accuracy of its estimated Pareto set, while simultaneously driving the front toward the true Pareto front. It can cope with noise whose characteristics change as a function of location (both design and objective), or which alter during the course of an optimization. Four state-of-the-art noise-tolerant EMO algorithms, as well as four widely used standard EMO algorithms, are compared to RTEA on 70 instances of ten continuous space test problems from the CEC’09 multiobjective optimization test suite. Different instances of these problems are generated by modifying them to exhibit different types and intensities of noise. RTEA seems to provide competitive performance across both the range of test problems used and noise types.

Comments Off

Enhancing Differential Evolution Utilizing Eigenvector-Based Crossover Operator

Differential evolution has been shown to be an effective methodology for solving optimization problems over continuous space. In this paper, we propose an eigenvector-based crossover operator. The proposed operator utilizes eigenvectors of covariance matrix of individual solutions, which makes the crossover rotationally invariant. More specifically, the donor vectors during crossover are modified, by projecting each donor vector onto the eigenvector basis that provides an alternative coordinate system. The proposed operator can be applied to any crossover strategy with minimal changes. The experimental results show that the proposed operator significantly improves DE performance on a set of 54 test functions in CEC 2011, BBOB 2012, and CEC 2013 benchmark sets.

Comments Off