Close
Mathematics of transfer learning and transfer risk: from medical to financial data analysis
31 May 2023, 4:00 pm–5:00 pm
The Women in Data Science and Mathematics Seminar Series welcomes Prof Xin Guo from UC Berkeley, followed by a Q&A session. All welcome to attend.
This event is free.
Event Information
Open to
- All
Availability
- Yes
Cost
- Free
Organiser
-
Professor Hao Ni
Transfer learning is an emerging and popular paradigm for utilizing existing knowledge from previous learning tasks to improve the performance of new ones. In this talk, we will first present transfer learning in the early diagnosis of eye diseases: diabetic retinopathy and retinopathy of prematurity. We will discuss how this empirical study leads to the mathematical analysis of the feasibility and transferability issues in transfer learning. We show how a mathematical framework for the general procedure of transfer learning helps establish the feasibility of transfer learning and for the analysis of the associated transfer risk.
Find more information on this event on our website, where you can also register for the seminar series.
Stay connected with us on LinkedIn or Twitter using #WINDSMATH.
About the Speaker
Prof Xin Guo
Prof Xin Guo is the Coleman Fung Chair professor at the college of engineering, UC Berkeley. Prior to that, she was an associate professor at the School of ORIE, Cornell University, and a research staff member at the Mathematics Department of IBM T.J. Watson Research Center. Her research interests are in stochastic processes, stochastic controls and games, and mathematics of machine learning, with applications to financial and medical data analysis. She is the co-editor-in-chief for Frontier of Mathematical Finance, and on the editorial board of several leading journals of controls, applied probability, mathematical finance, and operation research. On the application side, her work in eye disease diagnosis has led to free access for millions of people who can not afford regular eye exams; her work on early cancer detection using machine learning techniques laid the mathematical foundation for multiple early cancer detections recently approved by FDA.