報告題目：The automatic sparsity of kernel feature selection
報 告 人：阮豐 助理教授 美國西北大學統計系
報告時間：2023年7月4日 10:00
報告地點：數學樓第一報告廳
校内聯系方式：王培潔 wangpeijie@jlu.edu.cn
報告摘要：In this talk, we will describe a new sparsity-inducing technique based on minimization a family of kernels, terminologically called kernel-based feature selection. Unlike standard sparsification methods which rely on l_1 penalization, early stopping or post-processing, kernel feature selection achieves sparsity seemingly effortlessly in finite samples, bypassing the need of explicit regularizations. We will shed light on our current theoretical understanding of this empirically observed phenomenon that pervasively happens. As an application of this finding, we will illustrate its potential use in constructing new algorithms consistent for feature selection in nonparametric settings.
This is based on joint work with Michael I. Jordan, and Keli Liu.
報告人簡介：Feng Ruan is currently an assistant professor at Department of Statistics and Data Science from Northwestern University. Previously, he obtained his Ph.D. in Statistics at Stanford, advised by John Duchi, and was a postdoctoral researcher in EECS at the University of California, Berkeley, advised by Professor Michael Jordan. His current research has three driving goals: (1) Build optimal statistical inferential procedures accounting for crucial resource constraints such as computation, privacy, etc. (2) Develop modeling and analytic tools that give a calculus for understanding generally solvable non-convex problems. (3) Design new objectives so that local algorithms achieve guaranteed performances for problems of combinatorial structures.
