毛晓军 | Robust Reduced Rank Regression in a Distributed Setting

时间:2019年10月28(周一)下午 15:00-16:00

地点:中北校区理科大楼A1514

题目:Robust Reduced Rank Regression in a Distributed Setting

报告人:毛晓军 复旦大学大数据学院 助理教授

主持人:方方 副教授

摘要:

This paper studies the reduced rank regression problem, which assumes a low-rank structure of the coefficient matrix, together with heavy-tailed noises. To address the heavy-tailed noise, we adopt the quantile loss function instead of commonly used squared loss. However, the non-smooth quantile loss brings new challenges to both computation and the development of statistical properties, especially when the data is large in size and distributed across different machines. To this end, we first transform the response variable and reformulate the problem into a trace-norm regularized least-squares problem, which greatly facilitates the computation. Based on this formulation, we further develop a distributed algorithm. Theoretically, we establish the convergence rate of the obtained estimator and the theoretical guarantee for rank recovery. The simulation analysis is provided to demonstrate the effectiveness of our method.

个人简介:

Xiaojun Mao is an assistant professor of the School of Data Science at Fudan University. He received his PhD from Iowa State University in 2018. Dr. Mao's research interests include Matrix Completion, Recommender Systems, Regularization methods (e.g.  ℓ1, ℓ2 and nuclear-norm penalty), Distributed Inference and Genomic Prediction.


发布者:张瑛发布时间:2019-10-09浏览次数:61