paper-with-me

홈 › Papers

Empirical study of the modulus as activation function in computer vision applications

2023-01-15 · Iván Vallés-Pérez, Emilio Soria-Olivas, Marcelino Martínez-Sober, Antonio J. Serrano-López, Joan Vila-Francés, Juan Gómez-Sanchís

In this work we propose a new non-monotonic activation function: the modulus. The majority of the reported research on nonlinearities is focused on monotonic functions. We empirically demonstrate how by using the modulus activation function on computer vision tasks the models generalize better than with other nonlinearities - up to a 15% accuracy increase in CIFAR100 and 4% in CIFAR10, relative to the best of the benchmark activations tested. With the proposed activation function the vanishing gradient and dying neurons problems disappear, because the derivative of the activation function is always 1 or -1. The simplicity of the proposed function and its derivative make this solution specially suitable for TinyML and hardware applications.

📄 PDF Abstract BibTeX arXiv:2301.05993

Code (1)

ivallesp/abs 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Learning Large-Scale Modular Addition with an Auxiliary Modulus

2026-05-08 · Hanato Kikuchi, Ryosuke Masuya, Kazuhiko Kawamoto, Hiroshi Kera arxiv

Learning parity functions, more general modular addition, is a challenging machine learning task due to its input sensitivity. A recent study substantially scaled modular addition learning in both the number of summands …

Neural Network Approximation: Three Hidden Layers Are Enough

2020-10-25 · Zuowei Shen, Haizhao Yang, Shijun Zhang

A three-hidden-layer neural network with super approximation power is introduced. This network is built with the floor function ($\lfloor x\rfloor$), the exponential function ($2^x$), the step function ($1_{x\geq 0}$), o…

KANs for Computer Vision: An Experimental Study

2024-11-27 · Karthik Mohan, Hanxiao Wang, Xiatian Zhu

This paper presents an experimental study of Kolmogorov-Arnold Networks (KANs) applied to computer vision tasks, particularly image classification. KANs introduce learnable activation functions on edges, offering flexibl…

image-classificationImage ClassificationKolmogorov-Arnold Networks

Convergence Analysis of Max-Min Exponential Neural Network Operators in Orlicz Space

2025-08-14 · Satyaranjan Pradhan, Madan Mohan Soren arxiv

In this current work, we propose a Max Min approach for approximating functions using exponential neural network operators. We extend this framework to develop the Max Min Kantorovich-type exponential neural network oper…

Tuning the activation function to optimize the forecast horizon of a reservoir computer

2023-12-20 · Lauren A. Hurley, Juan G. Restrepo, Sean E. Shaheen

Reservoir computing is a machine learning framework where the readouts from a nonlinear system (the reservoir) are trained so that the output from the reservoir, when forced with an input signal, reproduces a desired out…

Hyperparameter Optimization