Sperf: a trainable smooth non-monotonic activation function with enhanced nonlinearity

DOI10.3934/MFC.2025037MaRDI QIDQ6848218zbMATH OpenFDO

Authors Xueying Li, Moyan Cao, Junli Zhao, Weibo Wei, Zhenkuan Pan

Publication date 6 March 2026

Published in Mathematical Foundations of Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.3934/MFC.2025037

zbMATH Keywords

activation functiondeep learningneural network

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07)

Cites work

• A mathematical explanation of UNet
• A trainable variational Chan-Vese network based on algorithm unfolding for image segmentation
• Approximation by superpositions of a sigmoidal function
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• Learning and approximating piecewise smooth functions by deep sigmoid neural networks
• Multilayer feedforward networks are universal approximators
• ReLU-like non-monotonic smooth activation functions based on regularized Heaviside functions and extensions
• SignReLU neural network and its approximation ability
• Smoothing piecewise linear activation functions based on mollified square root functions

This page was built for publication: Sperf: a trainable smooth non-monotonic activation function with enhanced nonlinearity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6848218)