南京大学计算机软件新技术国家重点实验室
摘要：
In the past decade, we have seen
dramatic evolution of machine learning from convex methods to non-convex
methods and from shallow models to deep models. Lying at the heart of machine
learning, mathematical optimization has played an important and indispensable
role in solving many different learning problems. However, there is still a big
gap between the practice used in deep learning community and the existing
theory. In this talk, I will focus on a learning paradigm called stagewise
learning that is different from conventional learning methods based on
stochastic gradient descent with a continuously decreasing step size. I will
show that the proposed stagewise learning algorithms can achieve
significant improvements in both theory and practice over standard stochastic
gradient method for solving many machine learning problems, including support
vector machine, AUC optimization, deep neural networks and generative
adversarial networks.
报告人简介:
Dr. Tianbao Yang
is currently an assistant professor at the University of Iowa. He received his
Ph.D. degree in Computer Science from Michigan State University in 2012. Before
joining UIowa, he
was a researcher in NEC Laboratories America at Cupertino (2013-2014) and a
Machine Learning Researcher in GE Global Research (2012-2013), mainly focusing
on developing deep learning algorithms and distributed optimization systems for
machine learning applications. Dr. Yang has board interests in machine learning
and he has focused on several research topics, including online learning,
distributed optimization, stochastic optimization, deep learning, and learning
theory. His recent research interests revolve around designing fast
optimization algorithms for non-convex problems including deep learning, GAN,
etc. He has published over 80 papers in top-notch venues. He has won multiple
awards including NSF Career Award, UIowa Dean’s Excellence Scholar Award,
the Best student paper award at 25th Conference on Learning Theory (COLT). He
is an associate editor of the Neurocomputing
Journal.
时间：5月16日（星期四）14:00
地点：计算机科学技术楼230室
|
