Report title: Multi-task Learning in Vector-valued Reproducing Kernel Banach Spaces with the l1 Norm

Reporter: Dr. Lin Rongrong Guangdong University of Technology

Reporting time: 10:45-11:20 am, September 17, 2020

Report location: Tencent Conference ID: 206 372 412

Conference password: 0917

School contact: Wang Rui [email protected]

Report summary:

Targeting at sparse multi-task learning, we consider regularization models with an l1 penalty on the coefficients of kernel functions. In order to provide a kernel method for this model, we construct a class of vector-valued reproducing kernel Banach spaces with the l1 norm. The notion of multi-task admissible kernels is proposed so that the constructed spaces could have desirable properties including the crucial linear representer theorem. Such kernels are related to bounded Lebesgue constants of a kernel interpolation question. We study the Lebesgue constant of multi-task kernels and provide examples of admissible kernels. Furthermore, we present numerical experiments for both synthetic data and real-world benchmark data to demonstrate the advantages of the proposed construction and regularization models. This is a joint work with Prof. Guohui Song (ODU) and Haizhang Zhang (SYSU).

Brief introduction of the speaker:

Lin Rongrong received his PhD from the School of Mathematics, Sun Yat-Sen University in June 2017; He served as a distinguished associate researcher of Sun Yat-Sen University from July 2017 to July 2020; he joined the School of Applied Mathematics of Guangdong University of Technology in August 2020 and is now a lecturer. He studied at the University of Alberta in Canada for one year and a short-term academic visit to Odominio University in the United States for two months. His research direction is machine learning kernel function method and time-frequency signal analysis, and he currently presides over the National Natural Science Youth Fund projects.