Joshua ChiengPARAMETRIC FLATTEN-T SWISH: AN ADAPTIVE NONLINEAR ACTIVATION FUNCTION FOR DEEP LEARNINGActivation function is a key component in deep learning that performs non-linear mappings between the inputs and outputs. Rectified Linear…Aug 23, 2021Aug 23, 2021
Joshua ChiengFlatten-T Swish: A Thresholded ReLU-Swish-like Activation Function for Deep LearningActivation functions are essential for deep learning methods to learn and perform complex tasks such as image classification. Since the…Aug 1, 2018Aug 1, 2018