English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Convolutional neural network (CNN) is one of the most frequently used deeplearning techniques. Various forms of models have been proposed and im-provedfor learning at CNN. When learning with CNN, it is necessary to determine theoptimal hyperparameters. However, the number of hyperparameters is so largethat it is difficult to do it manually, so much research has been done onautomation. A method that uses metaheuristic algorithms is attracting attentionin research on hyperparameter optimization. Metaheuristic algorithms arenaturally inspired and include evolution strategies, genetic algorithms,antcolony optimization and particle swarm optimization. In particular, particleswarm optimization converges faster than genetic algorithms, and various modelshave been proposed. In this paper, we pro-pose CNN hyperparameter optimizationwith linearly decreasing weight particle swarm optimization (LDWPSO). In theexperiment, the MNIST data set and CIFAR-10 data set, which are often used asbenchmark data sets, are used. By opti-mizing CNN hyperparameters with LDWPSO,learning the MNIST and CIFAR-10 datasets, we compare the accuracy with astandard CNN based on LeNet-5. As a result, when using the MNIST dataset, thebaseline CNN is 94.02% at the 5th epoch, compared to 98.95% for LDWPSO CNN,which improves accuracy. When using the CIFAR-10 dataset, the Baseline CNN is28.07% at the 10th epoch, compared to 69.37% for the LDWPSO CNN, which greatlyimproves accuracy. This paper is presented at the 36th Annual Conference of theJapanese Society for Artificial In-telligence. The final version is availableat the following URL: https://doi.org/10.11517/pjsai.JSAI2022.0_2S4IS2b03

Optimization of Convolutional Neural Network Using the Linearly Decreasing Weight Particle Swarm Optimization