# Dynamic ReLU

Source: [Dynamic ReLU](<https://www.sota2.com/research/paper/68fa68181e209131cea1a314>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:29:00.055Z

Authors: Yinpeng Chen, Xiyang Dai, Mengchen Liu, Dongdong Chen, Lu Yuan, Zicheng Liu

Published: 2020-03-22T23:45:35.000Z

## Abstract

Rectified linear units (ReLU) are commonly used in deep neural networks. So far ReLU and its generalizations (non-parametric or parametric) are static, performing identically for all input samples. In this paper, we propose dynamic ReLU (DY-ReLU), a dynamic rectifier of which parameters are generated by a hyper function over all in-put elements. The key insight is that DY-ReLU encodes the global context into the hyper function, and adapts the piecewise linear activation function accordingly. Compared to its static counterpart, DY-ReLU has negligible extra computational cost, but significantly more representation capability, especially for light-weight neural networks. By simply using DY-ReLU for MobileNetV2, the top-1 accuracy on ImageNet classification is boosted from 72.0% to 76.2% with only 5% additional FLOPs.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2003.10027>)

[PDF](<https://arxiv.org/pdf/2003.10027.pdf>)