Global-best optimization of ANN trained by PSO using the non-extensive cross-entropy with Gaussian gain

Susan, Seba; Ranjan, Rohit; Taluja, Udyant; Rai, Shivang; Agarwal, Pranav

doi:10.1007/s00500-020-05080-7

Cited by 5 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the DNN ensemble instead of a single DNN also minimizes, to some extent, the problem of convergence to local minima due to gradient descent optimization [3,4]. Alternative solutions to usage of ensembles for inducing variety in learning include the global optimization of network weights using evolutionary algorithms such as Particle Swarm Optimization [5].…”

Section: Introductionmentioning

confidence: 99%

Evaluating Deep Neural Network Ensembles by Majority Voting Cum Meta-Learning Scheme

Jain

Kumar

Susan

2021

Advances in Intelligent Systems and Computing

Self Cite

View full text Add to dashboard Cite

Deep Neural Networks (DNNs) are prone to overfitting and hence have high variance. Overfitted networks do not perform well for a new data instance. So instead of using a single DNN as classifier we propose an ensemble of seven independent DNN learners by varying only the input to these DNNs keeping their architecture and intrinsic properties same. To induce variety in the training input, for each of the seven DNNs, one-seventh of the data is deleted and replenished by bootstrap sampling from the remaining samples. We have proposed a novel technique for combining the prediction of the DNN learners in the ensemble. Our method is called pre-filtering by majority voting coupled with stacked meta-learner which performs a two-step confidence check for the predictions before assigning the final class labels. All the algorithms in this paper have been tested on five benchmark datasets namely, Human Activity Recognition (HAR), Gas sensor array drift, Isolet, Spambase and Internet advertisements. Our ensemble approach achieves higher accuracy than a single DNN and the average individual accuracies of DNNs in the ensemble, as well as the baseline approaches of plurality voting and meta-learning.

show abstract

Section: Introductionmentioning

confidence: 99%