欢迎光临!
您现在所在的位置:首页 >> 通知公告 & 学术信息
学术信息
SEMINARS
Deep learning: an historical perspective
时间  Datetime
2018-10-31 14:30 — 15:30 
地点  Venue
0
报告人  Speaker
Stéphane Canu
单位  Affiliation
法国应用科学院
邀请人  Host
牛一帅
报告摘要  Abstract

Deep learning may be one of the most overhyped of modern technologies, but there is a good chance that it will one day become the secret sauce in many different business processes". This presentation is a brief history of neural nets and deep learning trying to explain why and how did we get there. We will go through new developments in deep learning covering basic motivations, ideas, models and optimization, identifying challenges and opportunities. It will focus on issues related with large scale learning that is: high dimensional features, large variety of visual classes, and large number of examples. Some recent advances in deep automatic architecture design and adversarial learning (GAN) will be also discussed.