ALReLU: A different approach on Leaky ReLU activation function to improve Neural Networks Performance
This paper addresses the 'dying ReLU problem' for neural network practitioners, offering an incremental improvement to existing activation functions.
This paper proposes Absolute Leaky ReLU (ALReLU), a modification of Leaky ReLU that uses the absolute values of its small negative gradient. This approach aims to resolve the 'dying ReLU problem' and demonstrates significant improvements over LReLU and ReLU in image, text, and tabular data classification tasks across five different datasets, including COVID-19 disease classification.
Despite the unresolved 'dying ReLU problem', the classical ReLU activation function (AF) has been extensively applied in Deep Neural Networks (DNN), in particular Convolutional Neural Networks (CNN), for image classification. The common gradient issues of ReLU pose challenges in applications on academy and industry sectors. Recent approaches for improvements are in a similar direction by just proposing variations of the AF, such as Leaky ReLU (LReLU), while maintaining the solution within the same unresolved gradient problems. In this paper, the Absolute Leaky ReLU (ALReLU) AF, a variation of LReLU, is proposed, as an alternative method to resolve the common 'dying ReLU problem' on NN-based algorithms for supervised learning. The experimental results demonstrate that by using the absolute values of LReLU's small negative gradient, has a significant improvement in comparison with LReLU and ReLU, on image classification of diseases such as COVID-19, text and tabular data classification tasks on five different datasets.