Matrik: Jurnal Manajemen, Teknik Informatika, dan Rekayasa Komputer Vol. 23, No. 2, March 2024, pp. 343∼352 ISSN: 2476-9843, accredited by Kemenristekdikti, Decree No: 200/M/KPT/2020 DOI: 10.30812/matrik.v23i2.3763 ❒ 343 Improving Convolutional Neural Networks Performance Using Modified Pooling Function Achmad Lukman 1 , Wahju Tjahjo Saputro 2 , Erni Seniwati 3 1 Telkom University, Bandung, Indonesia 2 Universitas Muhammadiyah Purworejo, Purworejo, Indonesia 3 Universitas Amikom Yogyakarta, Yogyakarta, Indonesia Article Info Article history: Received January 08, 2024 Revised January 23, 2024 Accepted March 01, 2024 Keywords: Convolutional Neural Network Visual Geometry Group-16 Qmax pooling function Qavg pooling function ABSTRACT The Visual Geometry Group-16 (VGG16) network architecture, as part of the development of convo- lutional neural networks, has been popular among researchers in solving classification tasks, so in this paper, we investigated the number of layers to find better performance. In addition, we also proposed two pooling function techniques inspired by existing research on mixed pooling functions, namely Qmax and Qavg. The purpose of the research was to see the advantages of our method; we conducted several test scenarios, including comparing several modified network configurations based on VGG16 as a baseline and involving our pooling technique and existing pooling functions. Then, the results of the first scenario, we selected a network that can adapt well to our pooling technique, which was then carried out several tests involving the Cifar10, Cifar100, TinyImageNet, and Street View House Num- bers (SVHN) datasets as benchmarks. In addition, we were also involved in several existing methods. The experiment results showed that Net-E has the highest performance, with 93.90% accuracy for Cifar10, 71.17% for Cifar100, and 52.84% for TinyImageNet. Still, the accuracy was low when the SVHN dataset was used. In addition, in comparison tests with several optimization algorithms using the Qavg pooling function, it can be seen that the best accuracy results lie in the SGD optimization algorithm, with 89.76% for Cifar10 and 89.06% for Cifar100. Copyright c 2024 The Authors. This is an open access article under the CC BY-SA license. Corresponding Author: Achmad Lukman, +6281342352282, Faculty of Informatics, Department of Information Technology, Telkom University, Bandung, Indonesia, Email: alukman@telkomuniversity.ac.id How to Cite: A. Lukman, W. Saputro, and E. Seniwati, ”Improving Convolutional Neural Networks Performance Using Modified Pooling Func- tion”, MATRIK: Jurnal Manajemen, Teknik Informatika, dan Rekayasa Komputer, Vol. 23, No. 2, pp. 343-352, Mar, 2024. This is an open access article under the CC BY-SA license (https://creativecommons.org/licenses/by-sa/4.0/ ) Journal homepage: https://journal.universitasbumigora.ac.id/index.php/matrik