Comparison of Applicability of Synergetic Models with Convolutional Neural Networks as an Example of Pattern Recognition
Julia A. Vishnevskaya, Bogdan Salyp · 2023
This dominant approach in contemporary science and industry-level applications is multilayer deep learning. In this article we compare convolutional neural networks (CNN) architecture with synergetic models. The CIFAR10 image multiclass dataset was chosen as a main task to compare two algorithms between each other. A synergetic model and a convolutional neural network were tasked to assign one of 10 labels to each of CIFAR10 images using multiclass classification. Both models tackled a problem comparatively well, archiving 79% and 75% accuracy respectively. Synergetic model outperformed convolutional neural network both in metric values and in less execution time. Pros and cons of both models were addressed in the results section.