公开(公告)号：US10387773B2

公开(公告)日：2019-08-20

申请号：US14582059

申请日：2014-12-23

Applicant: eBay Inc.

Inventor： Zhicheng Yan ,  Robinson Piramuthu ,  Vignesh Jagadeesh ,  Wei Di ,  Dennis Decoste

IPC: G06N3/08 ,  G06N3/04

Abstract: Hierarchical branching deep convolutional neural networks (HD-CNNs) improve existing convolutional neural network (CNN) technology. In a HD-CNN, classes that can be easily distinguished are classified in a higher layer coarse category CNN, while the most difficult classifications are done on lower layer fine category CNNs. Multinomial logistic loss and a novel temporal sparsity penalty may be used in HD-CNN training. The use of multinomial logistic loss and a temporal sparsity penalty causes each branching component to deal with distinct subsets of categories.