Microsoft Research Surpasses Human-Level Performance On ImageNet Classification Dataset

Reading time icon 1 min. read


Readers help support MSPoweruser. When you make a purchase using links on our site, we may earn an affiliate commission. Tooltip Icon

Read the affiliate disclosure page to find out how can you help MSPoweruser effortlessly and without spending any money. Read more

Microsoft Research

Microsoft Research has recently published an academic paper titled “Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification”. In this paper, they are proposing a new rectifier model that surpasses human-level performance on visual recognition challenge.

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.

Download the full paper from the link below.

Source: Cornell University

More about the topics: Classification, GoogLeNet, Image Recognition, ImageNet, microsoft, Neural Networks, Rectified

Leave a Reply

Your email address will not be published. Required fields are marked *