设为首页收藏本站

VALSE

VALSE 首页 活动通知 研究组巡讲 查看内容

20150312-07 屠卓文:Deeply-Supervised Nets

2015-3-15 16:53| 发布者: zhenghaiyong| 查看: 5683| 评论: 0|来自: VALSE

摘要: 主题:Deeply-Supervised Nets主讲人:屠卓文(加州大学圣地亚哥分校)主持人:山世光(中科院计算所)活动时间:2015年03月12日,北京时间13:00-14:00。报告摘要dsn.pdfIn this talk, we will mainly discuss our r ...
主题:Deeply-Supervised Nets
主讲人:屠卓文(加州大学圣地亚哥分校)
主持人:山世光(中科院计算所)

活动时间:2015年03月12日,北京时间13:00-14:00。

  • In this talk, we will mainly discuss our recent deeply-supervised nets (DSN) work (at the NIPS14 deep learning workshop and also at AISTATS 2015 http://pages.ucsd.edu/~ztu/publication/aistats15_dsn.pdf), a method that simultaneously minimizes classification error and improves the directness and transparency of the hidden layer learning process. We focus our attention on three aspects of traditional convolutional-neural-network-type (CNN-type) architectures: (1) transparency in the effect intermediate layers have on overall classification; (2) discriminativeness and robustness of learned features, especially in early layers; (3) training effectiveness in the face of “vanishing” gradients. To combat these issues, we introduce “companion” objective functions at each hidden layer, in addition to the overall objective function at the output layer (an integrated strategy distinct from layer-wise pre-training). We also analyze our algorithm using techniques extended from stochastic gradient methods. The advantages provided by our method are evident in our experimental results, showing state-of-the-art performance on MNIST, CIFAR-10, CIFAR-100, and SVHN.
报告人简介
  • 屠老师就不用介绍了。一些如雷贯耳的大作,如DDMCMC,Image Parsing,Auto-context等都是屠老师的大作。详见: 屠老师主页和 http://scholar.glgoo.com/citations?user=9oz-dvgAAAAJ&hl=zh-CN&oi=ao

最新评论

Archiver|手机版|小黑屋|Vision And Learning SEminar    

GMT+8, 2020-8-3 23:28 , Processed in 0.029801 second(s), 20 queries .

Powered by Discuz! X3.2

© 2001-2013 Comsenz Inc.

返回顶部