Optimizing neural network loss function with differential evolution : доклад, тезисы доклада

Morozov Eduard; Stanovov Vladimir; Gorbunov Sergey

Дополнительные ссылки: eLibrary · Источник статьи

База данных: Каталог библиотеки СФУ

Библиографическое описание: Optimizing neural network loss function with differential evolution : доклад, тезисы доклада / E. Morozov, V. Stanovov, S. Gorbunov. - [S. l. : s. n.], 2026. - Текст : непосредственный // Proceedings of the IV International Conference on Advances in Materials, Earth Science and Technology (CAMSTech-IV-2024). - Melville, 2026. - P. 60026. - This work was funded by the Ministry of Science and Higher Education of the Russian Federation, State Contract FEFE-2023-0004. - DOI 10.1063/5.0322146.

Аннотация: In this paper, a method for optimizing the parameters of the loss function for the classification problem by neural networks is proposed. A simplified Taylor polynomial was used as the loss function. Differential evolution with success rate-based adaptation was used to find the optimal parameters. The cross-entropy value was used as the fitness value. Experiments on CIFAR-10 task can show, the purposed approach allows to achieve better results compared using Cross-Entropy. The curve of the discovered loss function exhibits a distinctly different form compared to the cross-entropy, featuring two distinct points of minimum at the probability of correct prediction equal to 0 and 1.

Год издания: 2026

Авторы: Morozov Eduard , Stanovov Vladimir , Gorbunov Sergey

Издательство: Melville

Источник: Proceedings of the IV International Conference on Advances in Materials, Earth Science and Technology (CAMSTech-IV-2024)

Номера страниц: 60026

Держатель оригинала документа: Reshetnev Siberian State University of Science and Technology, Siberian Federal University

Мероприятие: IV International Conference on Advances in Materials, Earth Science and Technology (CAMSTech-IV-2024) (Bukhara, 2024 ; 29.10 - 31.10)

Количество экземпляров:

Твёрдая копия издания отсутствует в фонде библиотеки

Показать дополнительные сведения

Multi-objective parametric synthesis of loss functions in neural networks with evolutionary… : доклад, тезисы доклада

Morozov Eduard и др.
2024 год

Evolutionary search algorithm for learning activation function of an artificial neural network : доклад, тезисы доклада

Yurshin Viacheslav
2025 год

Optimizing Convolutional Neural Network Architecture for Microscopy Image Recognition for Tuberculosis Diagnosis :…

Udegova E.S. и др.
2022 год

Neural network-based vehicle control in simulated environments using real-coded genetic… : доклад, тезисы доклада

Shabalin Denis и др.
2025 год

Adaptive Differential Evolution with Two Populations of New and Best Individuals : доклад, тезисы доклада

Stanovov Vladimir и др.
2022 год

Visualizing Parameter Adaptation in Differential Evolution with Expected Fitness Improvement : доклад, тезисы доклада

Stanovov V. и др.
2020 год

Analysis of the Effectiveness of Neural Networks with Different Configurations : доклад, тезисы доклада

Degtyareva Ksenia и др.
2024 год

Osteoarthritis Knee MRI Classification Using Convolutional Neural Networks : доклад, тезисы доклада

Lukyanova Natalia и др.
2026 год

Neural network controller identification for refining process : доклад, тезисы доклада, статья из сборника материалов…

Bukhtoyarov V.V. и др.
2019 год

Adaptive component crossover for differential evolution in solving single-objective… : доклад, тезисы доклада

Sopov Anton и др.
2025 год

Multi-Criteria Genetic Algorithm for Optimizing Distributed Computing Systems in Neural Network… : научное издание

Tynchenko Valeriya V. и др.
2025 год

On identification of neural correlates of reflection in simple recurrent neural networks : доклад, тезисы доклада,…

Bartsev S. и др.
2021 год

Searching for Multimode Resonator Topologies with Adaptive Differential Evolution : научное издание

Stanovov Vladimir и др.
2025 год

Neural networks playing ‘matching pennies' with each other: reproducibility of game dynamics : доклад, тезисы доклада

Dolgova T и др.
2019 год

Computing the dynamic AC of an electrical network via a fuzzy adaptive recurrent neural network : доклад, тезисы доклада

Mourtas Spyridon D.
2025 год

Ganetic loss for generative adversarial networks with a focus on medical applications : научное издание

Akhmedova Shakhnaz и др.
2025 год

Библиотечно-издательский комплекс СФУ

Optimizing neural network loss function with differential evolution : доклад, тезисы доклада
доклад, тезисы доклада, статья из сборника материалов конференций

Похожие публикации