Improving Learning of Neural Networks for Classification, Segmentation, and Associative Memory

Autor:	Knoblauch, Andreas, Luniak, Philipp
Jazyk:	angličtina
Rok vydání:	2022
Předmět:	Learning schedule Neural networks (Computer science) Cross-entropy method Neuronales Netz Cross-Entropie-Methode DDC 620 / Engineering & allied operations Klassifikation Cross entropy Power error loss ddc:620 Classification
Popis:	Machine learning models for classification and segmentation typically use loss functions based on binary or categorical cross entropy (BCE/CCE). Recent work has pointed out that cross entropy is not always optimal for maximizing accuracy of classification models. Instead, power error loss has been suggested as a generalization of BCE employing a power error exponent parameter q that can be adapted to output error distributions. In this paper we apply the power error loss to classification and segmentation using Convolutional Neural Network and a variant of DeepLabV3+. In line with previous findings, classification and segmentation performance can be improved significantly by using an appropriate fixed q. While previous works have found improvements mainly for q>1, here we report also cases where optimal q is publishedVersion
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d99c5b13dff6dc7d4518087fd6c57519 Zobrazit plný text záznamu