Designed Dithering Sign Activation for Binary Neural Networks

Abstract--Binary Neural Networks emerged as a cost-effective and energy-efficient solution for computer vision tasks by binarizing either network weights or activations. This work proposes an activation that applies multiple thresholds following dithering principles, shifting the Sign activation function for each pixel according to a spatially periodic threshold kernel. Unlike literature methods, the shifting is defined jointly for a set of adjacent pixels, taking advantage of spatial correlations. Experiments over the classification task demonstrate the effectiveness of the designed dithering Sign activation function as an alternative activation for binary neural networks, without increasing the computational cost. DNNs usually operate representation and preserve the precision. Binary neural networks (BNNs) connote an is a recurrent phenomenon in binary image representation, alternative that applies binarization strategies over the architecture mitigated through dithering strategies that adjust the density parameters, including weights [5], activations [6], or of binary values in the output image to closely approximate both [7] to handle the complexity.