Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
This work solves the problem of training deep neural networks more effectively for computer vision researchers, representing a significant advance rather than an incremental improvement.
The authors tackled image classification on ImageNet by proposing Parametric Rectified Linear Units (PReLU) and a robust initialization method, achieving a top-5 test error of 4.94%, which surpasses human-level performance and improves over the previous best by 26%.
Rectified activation units (rectifiers) are essential for state-of-the-art neural networks. In this work, we study rectifier neural networks for image classification from two aspects. First, we propose a Parametric Rectified Linear Unit (PReLU) that generalizes the traditional rectified unit. PReLU improves model fitting with nearly zero extra computational cost and little overfitting risk. Second, we derive a robust initialization method that particularly considers the rectifier nonlinearities. This method enables us to train extremely deep rectified models directly from scratch and to investigate deeper or wider network architectures. Based on our PReLU networks (PReLU-nets), we achieve 4.94% top-5 test error on the ImageNet 2012 classification dataset. This is a 26% relative improvement over the ILSVRC 2014 winner (GoogLeNet, 6.66%). To our knowledge, our result is the first to surpass human-level performance (5.1%, Russakovsky et al.) on this visual recognition challenge.