Deep Learning:A new wave of big data machine learning era
Kai Yu
Professor
deputy engeering director of multimedia department,Baidu
Abstract: In the past 30 years, tremendous progress has been achieved in building effective shallow classification models. Despite the success, we have to realize that, in major AI challenges, the key bottleneck is not the qualify of classifiers but that of features. Since 2006, learning high-level features using deep architectures from raw data has become a huge wave of new learning paradigms. In recent two years, various performance breakthroughs were repeatedly reported in image understanding, speech recognition, and bioinformatics areas, suggesting deep learning is not something ignorable for machine learning research. In this talk, I will walk through the recent history and key building blocks of deep learning, and discuss the major research topics, e.g., modeling and scalability issues. In the end, I will discuss what might be interesting for future research.
Bio: 余凯博士是百度公司技术副总监，毕业于南京大学，在德国慕尼黑大学获博士学位，曾在西门子德国总部任Senior Scientist，和NEC美国研究院任Department Head。他是机器学习领域的知名专家，拥有几十项专利和专利申请，发表了70多篇论文，被引用3000次以上，H-index为30。他是众多顶级国际会议（如NIPS, ICML, SIGIR, KDD, IJCAI等）程序委员会成员，并担任多个国际顶级杂志（如JMLR, TPAMI, MLJ, TNN, TKDE等）和美国自然科学基金NSF的专家评委，以及顶尖机器学习国际会议ICML和NIPS的领域主席（Area Chair）。2011年应邀在著名的斯坦福大学计算机系任Adjunct Faculty，主讲研究生课程“CS121: Introduction to AI”。他领导的NEC团队在PASAL VOC, ImageNet等竞赛上多次获得国际第一。在百度他领导语音，图像等基础技术团队，其中语音技术团队获得仅有的三个“2012年度百度最佳团队"之一。他推动了深度学习在百度的各个互联网业务方向的技术研发和应用。他获得“中关村高端领军人才”和"北京市海聚计划海外高层次人才”荣誉，入选第9批中组部“千人计划”国家特聘专家。他是北京邮电大学兼职教授和中科院计算所客座研究员。