[2S4-IS-2b] Machine learning

[2S4-IS-2b-03] Optimization of Convolutional Neural Network Using the Linearly Decreasing Weight Particle Swarm Optimization

〇Tatsuki Serizawa^1,2, Hamido Fujita¹ (1. Iwate Prefectural University, 2. Welmo Inc.)

Regular

キーワード：Convolutional Neural Network, Hyperparameter Optimization, Particle Swarm Optimization, MNIST, CIFAR-10

Convolutional neural network (CNN) is one of the most frequently used deep learning techniques. When learning with CNN, it is necessary to determine the optimal hyperparameters. A method that uses metaheuristic algorithms is attracting attention in research on hyperparameter optimization. In particular, particle swarm optimization converges faster than genetic algorithms, and various models have been proposed. In this paper, we propose CNN hyperparameter optimization with linearly decreasing weight particle swarm optimization (LDWPSO). In the experiment, by optimizing CNN hyperparameters with LDWPSO, learning the MNIST and CIFAR-10 datasets, we compare the accuracy with a standard CNN based on LeNet-5. As a result, when using the MNIST dataset, the baseline 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 is 26.45% at the 9th epoch, compared to 69.53% for the LDWPSO CNN, which greatly improves accuracy.