Empirical Review of Standard Benchmark Functions Using Evolutionary Global Optimization

Dieterich, Johannes M.; Hartke, Bernd

doi:10.4236/am.2012.330215

Cited by 90 publications

(45 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Eight standard testing benchmark functions were compared experimentally and the results are listed in Table 1 (Bratton & Kennedy, 2007;Dieterich & Hartke, 2012;Geng et al, 2013;Yang & Wang, 2009). In the experiment, the population size is 15.…”

Section: Experimental Designmentioning

confidence: 99%

A cautious PSO with conditional random

Chan

Chen

2015

Expert Systems with Applications

View full text Add to dashboard Cite

Section: Experimental Designmentioning

confidence: 99%

A cautious PSO with conditional random

Chan

Chen

2015

Expert Systems with Applications

View full text Add to dashboard Cite

“…Rastrigin's function has been used by several researchers as a hard benchmark function to test experimental optimisation algorithms (Dieterich and Hartke, 2012;Törn and Zilinskas, 1989;Mühlenbein et al, 1991;Liang, 2011;Liang et al, 2006;Ali et al, 2005). Here, if not stated otherwise, we consider default settings for PISAA: (i) n " 10 6 iterations, (ii) uniformly spaced grid tu j u with m " 400, u 1 "´0.01, u 400 " 40, (iii) desirable probability with parameter λ " 0.1, (iv) temperature ladder tτ t u with τ h " 1, n pτ q " 1, τ˚" 10´2, (iv) gain factor tγ t u with n pγq " 10 5 , β " 0.55.…”

Section: Rastrigin's Functionmentioning

confidence: 99%

Parallel and interacting stochastic approximation annealing algorithms for global optimisation

et al. 2016

View full text Add to dashboard Cite

We present the parallel and interacting stochastic approximation annealing (PISAA) algorithm, a stochastic simulation procedure for global optimisation, that extends and improves the stochastic approximation annealing (SAA) by using population Monte Carlo ideas. The standard SAA algorithm guarantees convergence to the global minimum when a square-root cooling schedule is used; however the efficiency of its performance depends crucially on its self-adjusting mechanism. Because its mechanism is based on information obtained from only a single chain, SAA may present slow convergence in complex optimisation problems. The proposed algorithm involves simulating a population of SAA chains that interact each other in a manner that ensures significant improvement of the self-adjusting mechanism and better exploration of the sampling space. Central to the proposed algorithm are the ideas of (i) recycling information from the whole population of Markov chains to design a more accurate/stable self-adjusting mechanism and (ii) incorporating more advanced proposals, such as crossover operations, for the exploration of the sampling space. PISAA presents a significantly improved performance in terms of convergence. PISAA can be implemented in parallel computing environments if available. We demonstrate the good performance of the proposed algorithm on challenging applications including Bayesian network learning and protein folding. Our numerical comparisons suggest that PISAA outperforms the simulated annealing, stochastic approximation annealing, and annealing evolutionary stochastic approximation Monte Carlo especially in high dimensional or rugged scenarios.

show abstract

“…In this paper, the GRUNGE (Gaussian: Randomized Uncorrelated Gaussian Extrema) method was employed to generate both the landscape and the moving peaks [19]. Let L denote the number of D-D landscape Gaussians, h l land the height of the l-th landscape Gaussian (l = 1, 2, …, L), u l the location of each Gaussian, and σ l its standard deviation (a separate value for each dimension).…”

Section: Cost Function: "Moving Peaks" Problem (32d and 64d)mentioning

confidence: 99%

Dynamic particle swarm optimization with heterogeneous multicore parallelism and GPU acceleration

Wachowiak

Wachowiak-Smolikova

Rotondo

2015

2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE)

View full text Add to dashboard Cite

Global optimization of dynamic cost functions is important in many engineering applications. For these tasks, global optima change over time, or are greatly affected by dynamic noise. Nature-based stochastic methods, including genetic algorithms, particle swarm optimization (PSO), and differential evolution, have been particularly effective in dynamic optimization. However, these methods are generally very computationally intensive, and consequently research has focused on parallelization paradigms. In this paper, PSO approaches for dynamic optimization are analyzed for parallelization opportunities on relatively inexpensive, readilyavailable heterogeneous parallel graphics processing unit (GPU) and multicore hardware. A sophisticated adaptation of PSO-a multi-swarm technique proposed for dynamic problems-is parallelized in different ways at multiple levels. Experimental results on high-dimensional "moving-peaks" functions show that high speedups can be obtained through making use of different high-performance components on commodity hardware. Heterogeneous high-performance computing is proposed as a way to mitigate the time complexity of dynamic PSO adaptions.

show abstract

Empirical Review of Standard Benchmark Functions Using Evolutionary Global Optimization

Cited by 90 publications

References 32 publications

A cautious PSO with conditional random

A cautious PSO with conditional random

Parallel and interacting stochastic approximation annealing algorithms for global optimisation

Dynamic particle swarm optimization with heterogeneous multicore parallelism and GPU acceleration

Contact Info

Product

Resources

About