RepAct: The Re-parameterizable Adaptive Activation Function

TL;DR

RepAct introduces a re-parameterizable adaptive activation function that enhances lightweight neural networks' performance on edge devices, achieving significant accuracy improvements with minimal additional computational cost.

Contribution

It proposes a novel multi-branch, learnable adaptive activation function tailored for efficient edge AI, outperforming traditional functions in lightweight models.

Findings

Up to 7.92% accuracy improvement on MobileNetV3-Small for ImageNet100

Maintains computational complexity comparable to HardSwish

Enhances feature processing and interpretability in lightweight networks

Abstract

Addressing the imperative need for efficient artificial intelligence in IoT and edge computing, this study presents RepAct, a re-parameterizable adaptive activation function tailored for optimizing lightweight neural networks within the computational limitations of edge devices. By employing a multi-branch structure with learnable adaptive weights, RepAct enriches feature processing and enhances cross-layer interpretability. When evaluated on tasks such as image classification and object detection, RepAct notably surpassed conventional activation functions in lightweight networks, delivering up to a 7.92% accuracy boost on MobileNetV3-Small for the ImageNet100 dataset, while maintaining computational complexity on par with HardSwish. This innovative approach not only maximizes model parameter efficiency but also significantly improves the performance and understanding capabilities of lightweight neural networks, demonstrating its potential for real-time edge computing applications.