Maintaining regularity and generalization in data using the minimum description length principle and genetic algorithm: Case of grammatical inference

Pandey, Hari Mohan; Chaudhary, Ankit; Mehrotra, Deepti; Kendall, Graham

doi:10.1016/j.swevo.2016.05.002

Cited by 21 publications

(12 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiple simulations have been performed, and the average results are reported. An extensive control parameter tuning is done by Taguchi signal to noise ratio (SNR) method along with orthogonal matrix as done in [58] [59] [60]. Taguchi SNR is a log function of the desired output that serves as an objective function as shown in eq.…”

Section: Simulation Settingsmentioning

confidence: 99%

Particle swarm optimization based energy efficient clustering and sink mobility in heterogeneous wireless sensor network

2020

Self Cite

View full text Add to dashboard Cite

In a WSN, sensor node plays a significant role. Working of sensor node depends upon its battery's life. Replacements of batteries are found infeasible once they are deployed in a remote or unattended area. Plethora of research had been conducted to address this challenge, but they suffer one or the other way. In this paper, a particle swarm optimization (PSO) algorithm integrated with an energy efficient clustering and sink mobility ((PSO-ECSM) is proposed to deal with both cluster head selection problem and sink mobility problem. Extensive computer simulations are conducted to determine the performance of the PSO-ECSM. Five factors such as residual energy, distance, node degree, average energy and energy consumption rate (ECR) are considered for CH selection. An optimum value of these factors is determined through PSO-ECSM algorithm. Further, PSO-ECSM addresses the concern of relaying the data traffic in a multi-hop network by introducing sink mobility. PSO-ECSM's performances are tested against the state-of-the-art algorithms considering five performance metrics (stability period, network, longevity, number of dead nodes against rounds, throughput and network's remaining energy). Statistical tests are conducted to determine the significance of the performance. Simulation results show that the PSO-ECSM improves stability period, half node dead, network lifetime and throughput vis-à-vis ICRPSO by 24.8%, 31.7%, 9.8 %, and 12.2%, respectively.

show abstract

Section: Simulation Settingsmentioning

confidence: 99%

Particle swarm optimization based energy efficient clustering and sink mobility in heterogeneous wireless sensor network

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Identifying these parameter values is important or increases the probability of global solution. If these parameter values have not been tuned correctly, then it leads to premature convergence -a situation when diversity decreases over some generations [19][20][21]. Exploration and exploitation is the key for the success of any meta-heuristic search algorithm.…”

Section: Proposed Techniquementioning

confidence: 99%

“…The researcher most of the time struggles to find the suitable parameter value for the metaheuristic search algorithms-same is the case with HS. In our approach, to tune the correct value of HMCR and HMS, we incorporated an orthogonal array based approach and a Taguchi method that determines the signal to noise ratio, helps in finding the right combinations of parameters [19][20][21]. PAR and BW parameters play an important role in the convergence of the algorithm to find the optimal solution [14].…”

Section: Proposed Techniquementioning

confidence: 99%

“…The parameter tuning of control parameter have been performed by using orthogonal array with Taguchi SNR method. The experiment utilises the orthogonal array which provides reduced variance on optimal value of control parameters [19][20][21]. Taguchi's S/N ratio represents the log function of desired output which works as objective function for optimisation of control parameters.…”

Section: Parameter Tuningmentioning

confidence: 99%

See 1 more Smart Citation

Efficient parameter estimation of software reliability growth models using harmony search

2017

View full text Add to dashboard Cite

The primary challenge of software reliability growth model is to find the unknown model parameters that are used to validate on software failure dataset. Though, numerical estimation technique plays a vital role in parameter estimation of software reliability growth models, they are not optimal as they suffer from constraints sucha as sample size, biasing, and initialisation of parameters. In this study, a parameter estimation of software reliability growth model that utilises a variant of harmony search is proposed. Extensive experiments are conducted on seven different software datasets of varying complexity. A robust experimental setup is developed employing an orthogonal array and Taguchi method. Twofold performance comparisons are performed. First, the authors tested their proposed approach against Cuckoo search and numerical method (least square estimation) considering mean square error and Theil's statistics as a quality measure. Second, the authors applied statistical tests are performed that demonstrate the superiority of their approach over the others. The underlying motivation to conduct this study is to motivate researchers to utilise their approach for a better estimation of model parameters.

show abstract

“…A comprehensive work on parameter calibration was presented in [12], though the authors suggest that better results can be achieved through Evolutionary Algorithms (EAs). GAs have been found very effective in several areas including grammar inference [14] [18], time tabling [19]. Considering this view, we propose a hybrid deep learning mechanism which utilizes the merits of GAs to enhance Gradient Decent in backpropagation learning.…”

Section: Introductionmentioning

confidence: 99%

A Genetic Deep Learning Model for Electrophysiological Soft Robotics

Pandey

Windridge

2020

Soft Computing Applications

Self Cite

View full text Add to dashboard Cite

Deep learning methods are modeled by means of multiple layers of predefined set of operations. These days, deep learning techniques utilizing unsupervised learning for training neural networks layers have shown effective results in various fields. Genetic algorithms, by contrast, are search and optimization algorithm that mimic evolutionary process. Previous scientific literatures reveal that genetic algorithms have been successfully implemented for training three-layer neural networks. In this paper, we propose a novel genetic approach to evolving deep learning networks. The performance of the proposed method is evaluated in the context of an electrophysiological soft robot like system, the results of which demonstrate that our proposed hybrid system is capable of effectively training a deep learning network.

show abstract

Maintaining regularity and generalization in data using the minimum description length principle and genetic algorithm: Case of grammatical inference

Abstract: Maintaining regularity and generalization in data using the minimum description length principle and genetic algorithm: case of grammatical inference. Swarm and Evolutionary Computation, 31. pp. 11-23.

Cited by 21 publications

References 52 publications

Particle swarm optimization based energy efficient clustering and sink mobility in heterogeneous wireless sensor network

Particle swarm optimization based energy efficient clustering and sink mobility in heterogeneous wireless sensor network

Efficient parameter estimation of software reliability growth models using harmony search

A Genetic Deep Learning Model for Electrophysiological Soft Robotics

Contact Info

Product

Resources

About