Combinations of genetic algorithms and neural networks: a survey of the state of the art

Schaffer, J. David; Whitley, Darrell; Eshelman, Larry J.

doi:10.1109/cogann.1992.273950

Cited by 401 publications

(210 citation statements)

References 47 publications

Supporting

Mentioning

188

Contrasting

Unclassified

Order By: Relevance

“…First, NE methods that evolve neurons (or more generally functional units) have been demonstrated as exhibiting superior solutions for various controller design tasks [19,21]. Second, cooperative co-evolutionary methods that operate at the neuron level avoid the competing conventions problem [45] and population diversity is maintained thus reducing the chance of premature convergence (relatively high for NE methods that evolve complete ANNs) [29].…”

Section: State-of-the-artmentioning

confidence: 99%

Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task

2009

View full text Add to dashboard Cite

This article presents results from an evaluation of the collective neuro-evolution (CONE) controller design method. CONE solves collective behavior tasks, and increases task performance via facilitating emergent behavioral specialization. Emergent specialization is guided by genotype and behavioral specialization difference metrics that regulate genotype recombination. CONE is comparatively tested and evaluated with similar neuro-evolution methods in an extension of the multi-rover task, where behavioral specialization is known to benefit task performance. The task is for multiple simulated autonomous vehicles (rovers) to maximize the detection of points of interest (red rocks) in a virtual environment. Results indicate that CONE is appropriate for deriving sets of specialized rover behaviors that complement each other such that a higher task performance, comparative to related controller design methods, is attained in the multi-rover task.

show abstract

Section: State-of-the-artmentioning

confidence: 99%

Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task

2009

View full text Add to dashboard Cite

show abstract

“…The parameters for crossover rate and mutation rate were adapted primarily based on a large empirical study by [Schaffer et al, 1989]. We also incorporate "elitism" [Davis, 1991], in which the GA keeps track of the best fitness chromosome in the population.…”

Section: Methodsmentioning

confidence: 99%

A Genetic Algorithm for Minimizing Query Processing Time in Distributed Database Design: Total Time Versus Response Time

Song¹

2009

The KIPS Transactions:PartD

View full text Add to dashboard Cite

Query execution time minimization is an important objective in distributed database design. While total time minimization is an objective for On Line Transaction Processing (OLTP), response time minimization is for Decision Support queries. We formulate the sub-query allocation problem using analytical models and solve with genetic algorithm (GA). We show that query execution plans with total time minimization objective are inefficient from response time perspective and vice versa. The procedure is tested with simulation experiments for queries of up to 20 joins. Comparison with exhaustive enumeration indicates that GA produced optimal solutions in all cases in much less time.

show abstract

“…In a continuous searching space, real-coded genetic algorithm is more natural than binary-coded genetic algorithm when we make the optimization process of decision variables. In this study, the artificial neural networks are trained with binarycoded genetic algorithm and real-coded genetic algorithm that has been used for the pattern classification problems [17], [18]. The block diagram of the GA is depicted in Fig.…”

Section: Outlines Of Genetic Algorithmsmentioning

confidence: 99%

Performance Analysis of Neural Networks Training using Real Coded Genetic Algorithm

Sarangi¹,

Majhi²,

Panda³

2012

IJCA

View full text Add to dashboard Cite

Multilayer perceptrons (MLPs) are widely used for pattern classification and regression problems. Backpropagation (BP) algorithm is known technique in the training of multilayer perceptrons. However for its optimum training convergence, the learning and momentum parameters need to be tuned on trial and error method. Further, sometimes the backpropagation algorithm fails to achieve global convergence. To alleviate these problems we suggest a genetic algorithm based training for MLP network. Both binary coded and real coded genetic algorithm are used and a comparative training performance analysis has been studied. It is observed from simulation results that both the schemes outperform backpropagation algorithm and achieve global convergence. Further the real coded GA based training shows a faster convergence than binary coded GA based training. For simulation datasets are taken from UCI based machine learning repository. Hence real-coded genetic algorithm finds an alternative for back propagation based training algorithm.

show abstract

Combinations of genetic algorithms and neural networks: a survey of the state of the art

Cited by 401 publications

References 47 publications

Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task

Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task

A Genetic Algorithm for Minimizing Query Processing Time in Distributed Database Design: Total Time Versus Response Time

Performance Analysis of Neural Networks Training using Real Coded Genetic Algorithm

Contact Info

Product

Resources

About