学术报告（二）
报告题目：Sketching for Kronecker Product Regression and P-splines
报告人：刁怀安副教授 (东北师范大学)
报告时间：2020/11/14(周六)下午 15:40-16:40
报告形式：腾讯会议
会议ID：408 520 583
报告摘要：TensorSketch is an oblivious linear sketch introduced in (Pagh, 2013) and later used in (Pham and Pagh, 2013) in the context of SVMs for polynomial kernels. It was shown in (Avron et al., 2014) that TensorSketch provides a subspace embedding, and therefore can be used for canonical correlation analy- sis, low rank approximation, and principal component regression for the polynomial ker- nel. We take TensorSketch outside of the context of polynomials kernels, and show its utility in applications in which the underly- ing design matrix is a Kronecker product of smaller matrices. This allows us to solve Kro- necker product regression and non-negative Kronecker product regression, as well as regu- larized spline regression. Our main technical result is then in extending TensorSketch to other norms. That is, TensorSketch only provides input sparsity time for Kronecker product regression with respect to the 2-norm. We show how to solve Kronecker product re- gression with respect to the 1-norm in time sublinear in the time required for computing the Kronecker product, as well as for more general p-norms.
报告人简介：刁怀安，副教授，博士生导师，2007年博士毕业于香港城市大学，研究兴趣为数值代数、随机化算法、微分算子谱理论、散射问题，在Journal de Mathématiques Pures et Appliquées、Calculus of Variations and Partial Differential Equations、Communications in Partial Differential Equations、Mathematics of Computation，以及机器学习领域顶级会议NeurIPS 2019，等国际主流期刊发表科研论文40余篇。出版学术专著一本。据Mathscinet显示，所发表论文累计被引用210次；据Web of Science显示，所发表论文中单篇最高被引用65次。主持并完成多项国家自然科学基金与教育部博士点基金项目。曾多次受邀访问国内外高校进行合作研究与学术交流。