Optimizing performance of feedforward and convolutional neural networks through dynamic activation functions

Training algorithms in the domain of deep learning, have led to significant breakthroughs across diverse and subsequent domains including speech, text, images, and video processing. While the research around deeper network architectures, notably exemplified by ResNet's expansive 152-layer structures, has yielded remarkable outcomes, the exploration of computationally simpler shallow Convolutional Neural Networks (CNN) remains an area for further exploration. Activation functions, crucial in introducing non-linearity within neural networks, have driven substantial advancements. In this paper, we delve into hidden layer activations, particularly examining their complex piece-wise linear attributes. Our comprehensive experiments showcase the superior efficacy of these piece-wise linear activations over traditional Rectified Linear Units across various architectures. We propose a novel Adaptive Activation algorithm, AdAct, exhibiting promising performance improvements in diverse CNN and multilayer perceptron configurations, thereby presenting compelling results to support its usage.