Activation function optimization method: Learnable series linear units (LSLUs)

TL;DR

This paper introduces LSLU, a learnable activation function that adapts dynamically during training, improving neural network accuracy and speed on various datasets by increasing non-linearity.

Contribution

We propose LSLU, a novel learnable series linear activation function that simplifies networks and enhances generalization through dynamic parameter adjustment.

Findings

LSLU improves CIFAR100 accuracy by 3.17%.

LSLU accelerates training convergence.

Learnable parameters adapt during training to boost non-linearity.

Abstract

Effective activation functions introduce non-linear transformations, providing neural networks with stronger fitting capa-bilities, which help them better adapt to real data distributions. Huawei Noah's Lab believes that dynamic activation functions are more suitable than static activation functions for enhancing the non-linear capabilities of neural networks. Tsinghua University's related research also suggests using dynamically adjusted activation functions. Building on the ideas of using fine-tuned activation functions from Tsinghua University and Huawei Noah's Lab, we propose a series-based learnable ac-tivation function called LSLU (Learnable Series Linear Units). This method simplifies deep learning networks while im-proving accuracy. This method introduces learnable parameters {\theta} and {\omega} to control the activation function, adapting it to the current layer's training stage and improving the model's generalization. The principle is to increase non-linearity in each activation layer, boosting the network's overall non-linearity. We evaluate LSLU's performance on CIFAR10, CIFAR100, and specific task datasets (e.g., Silkworm), validating its effectiveness. The convergence behavior of the learnable parameters {\theta} and {\omega}, as well as their effects on generalization, are analyzed. Our empirical results show that LSLU enhances the general-ization ability of the original model in various tasks while speeding up training. In VanillaNet training, parameter {\theta} initially decreases, then increases before stabilizing, while {\omega} shows an opposite trend. Ultimately, LSLU achieves a 3.17% accuracy improvement on CIFAR100 for VanillaNet (Table 3). Codes are available at https://github.com/vontran2021/Learnable-series-linear-units-LSLU.