Brian W. Goldman's research while affiliated with Michigan State University and other places

Publications (19)

Conference Paper
Full-text available
This paper presents the results of the second edition of the Wind Farm Layout Optimization Competition, which was held at the 22nd Genetic and Evolutionary Computation COnference (GECCO) in 2015. During this competition, competitors were tasked with optimizing the layouts of five generated wind farms based on a simplified cost of energy evaluation...
Article
This paper presents the results of the second edition of the Wind Farm Layout Optimization Competition, which was held at the 22nd Genetic and Evolutionary Computation COnference (GECCO) in 2015. During this competition, competitors were tasked with optimizing the layouts of five generated wind farms based on a simplified cost of energy evaluation...
Article
The parameter-less population pyramid (P3) is a recent evolutionary computation algorithm proposed for black box optimization. Shown to be efficient for a variety of benchmark problems, P3 replaces the conventional constant population model with expanding sets of expanding populations. We investigated how this new metaheuristic optimization algorit...
Conference Paper
Runtime analysis of black-box search algorithms provides rigorous performance guarantees, aiding in algorithm design and comparison. Unfortunately, deriving bounds can be challenging and as a result existing literature has focused on simplistic algorithms. The Parameter-less Population Pyramid (P3) is a recently proposed (Goldman and Punch, GECCO 2...
Article
This paper investigates Gray Box Optimization for pseudo-Boolean optimization problems composed of M subfunctions, where each subfunction accepts at most k variables. We will refer to these as Mk Landscapes. In Gray Box Optimization, the optimizer is given access to the set of M subfunctions. We prove Gray Box Optimization can efficiently compute h...
Conference Paper
Examining the properties of local optima is a common method for understanding combinatorial-problem landscapes. Unfortunately, exhaustive algorithms for finding local optima are limited to very small problem sizes. We propose a method for exploiting problem structure to skip hyperplanes that cannot contain local optima, allowing runtime to scale wi...
Conference Paper
Unlike black-box optimization problems, gray-box optimization problems have known, limited, non-linear relationships between variables. Though more restrictive, gray-box problems include many real-world applications in network security, computational biology, VLSI design, and statistical physics. Leveraging these restrictions, the Hamming-Ball Hill...
Article
The Parameter-less Population Pyramid (P3) is a recently introduced method for performing evolutionary optimization without requiring any user specified parameters. P3's primary innovation is to replace the generational model with a pyramid of multiple populations that are iteratively created and expanded. In combination with local search and advan...
Article
Full-text available
Real world applications of evolutionary techniques are often hindered by the need to determine problem specific parameter settings. While some previous methods have reduced or removed the need for parameter tuning, many do so by trading efficiency for general applicability. The Parameter-less Population Pyramid (P3) is an evolutionary technique tha...
Article
Full-text available
Current theory suggests that many signaling systems evolved from preexisting cues. In aposematic systems, prey warning signals benefit both predator and prey. When the signal is highly beneficial, a third species often evolves to mimic the toxic species, exploiting the signaling system for its own protection. We investigated the evolutionary dynami...
Article
Understanding how search operators interact with solution representation is a critical step to improving search. In Cartesian genetic programming (CGP), and genetic programming (GP) in general, the complex genotype to phenotype map makes achieving this understanding a challenge. By examining aspects such as tuned parameter values, the search qualit...
Conference Paper
Creating an Evolutionary Algorithm (EA) which is capable of automatically configuring itself and dynamically controlling its parameters is a challenging problem. However, solving this problem can reduce the amount of manual configuration required to implement an EA, allow the EA to be more adaptable, and produce better results on a range of problem...
Conference Paper
In this paper we examine how Cartesian Genetic Programming's (CGP's) method for encoding directed acyclic graphs (DAGs) and its mutation operator bias the effective length of individuals as well as the distribution of inactive nodes in the genome. We investigate these biases experimentally using two CGP variants as comparisons: Reorder, a method fo...
Conference Paper
Cartesian Genetic Programming (CGP) is a form of Genetic Programming (GP) where a large proportion of the genome is identifiably unused by the phenotype. This can lead mutation to create offspring that are genotypically different but phenotypically identical, and therefore do not need to be evaluated. We investigate theoretically and empirically th...
Article
Full-text available
We present the results of a community survey regarding genetic programming (GP) benchmark practices. Analysis shows broad consensus that improvement is needed in problem selection and experimental rigor. While views expressed in the survey dissuade us from proposing a large-scale benchmark suite, we find community support for creating a blacklist o...
Conference Paper
Automatically configuring and dynamically controlling an Evolutionary Algorithm's (EA's) parameters is a complex task, yet doing so allows EAs to become more powerful and require less problem specific tuning to become effective. Supportive Coevolution is a new form of Evolutionary Algorithm (EA) that uses multiple populations to overcome the limita...
Conference Paper
Discovering and exploiting the linkage between genes during evolutionary search allows the Linkage Tree Genetic Algorithm (LTGA) to maximize crossover effectiveness, greatly reducing both population size and total number of evaluations required to reach success on decomposable problems. This paper presents a comparative analysis of the most promine...
Conference Paper
Traditional evolutionary algorithms (EAs) are powerful problem solvers that have several fixed parameters which require tuning. An increasing body of evidence suggests that the optimal values of some, if not all, EA parameters change during the course of executing an evolutionary run. This paper investigates the potential benefits of dynamic parame...
Conference Paper
Crossover is a core genetic operator in many evolutionary algorithms (EAs). The performance of such EAs on a given problem is dependent on properly configuring crossover. A small set of common crossover operators is used in the vast majority of EAs, typically fixed for the entire evolutionary run. Selecting which crossover operator to use and tunin...

Citations

... where Pyearly represents the yearly profit of the wind farm, M implies the marketing cost per unit of wind power, C symbolizes the generation expense per unit of wind power and Gyearly denotes the wind power generated per year. The generation expense of wind power has been calculated as per the cost function presented by an established function [7]. In this research work, the wind energy generation capacity of Jafrabad has been studied. ...
... The obtained results compared to the previously recorded results using different optimization algorithms. Gandomi and Goldman (2018) tried the parameter-less population pyramid (P3) for truss optimization with discrete design variables. As P3 is a black-box evolutionary optimization algorithm, the results were compared to some other well-known black-box algorithms, including random restart hill climbing (RRHC), parameter-less hierarchical Bayesian optimization algorithm (PHBOA), DE, and a modified GA. ...
... The theory of attractors (Garnier and Kallel 2001; Reeves and Eremeev 2004;Goldman and Punch 2016) and their empirical estimations (Ochoa et al. 2008;Verel et al. 2011) focus on single-objective functions. The methods for estimating the number of local optima are empirically compared on several combinatorial optimization problems in Hernando et al. (2013), Elorza et al. (2018). ...
... The function H-IFF (Hierarchical If and only If) [59] consists of hierarchical building blocks that need to attain equal values in order to contribute to the fitness. It was studied theoretically [9], [35] and is frequently used in empirical studies, see, e. g. [33], [58]. ...
... If a priori information about the function is accessible, it can significantly support the search and should be considered during the algorithm design. Current research on algorithm designs that include structural operators, such as function decomposition, is known as grey-box optimization (Whitley et al. 2016;Santana 2017). However, many modern algorithms focus on handling black-box problems where the problem includes little or no a priori information. ...
... A number of methods have been covered under the umbrella of gray-box optimizers. From highly efficient hill-climbers [21,175], to enhanced partition crossover operators [20,163], and combinations of black-box global optimizers and local-search gray-box optimizers [43]. ...
... The synthesis of FPTs by CGP can also be found in [35]. The authors apply the improvements in CGP proposed by [16] and implemented the well-known NSGA-II strategy to deal with two conflicting objectives, namely, the accuracy and the size of the tree. ...
... There exist multiple MBOAs that show state-of-the-art performance [16,22,36,46]. However, as we discuss in this paper, the model used to adapt the solution neighbourhood and the method used to then explore this neighbourhood differ between the algorithms. ...
... We note that a linear runtime of the (1 + (λ, λ)) GA on OneMax was obtained earlier with a self-adjusting choice of the mutation rate based on the one-fifth rule [8]. While this worked well on OneMax, experimental [17] and theoretical [7] studies on satisfiable MAX-3SAT instances showed that this approach carries the risk that the population size λ increases rapidly because the problem structure may just not allow a one-fifth success rate, regardless how large λ is. Since this behavior increases the time complexity of each iteration, it leads to a significant performance loss. ...
... This design choice was made with the eventual goal of evolving them and the ShoreFor calibrated parameters (inputs 3, 8 and 9). Input (4) is used in Equation 2, whereas inputs (5,6,7,8,9) are used at the end of the genome to compute the integral of Equation 1 such that the output of the CGP individual corresponds to the value of -(the cross-shore shoreline location). ...