报告人：胡婷
报告地点：腾讯会议
报告时间：2020年08月21日星期五09:30-10:30
邀请人：刘红
报告摘要：
Distributed learning has received increasing attention in recent years for its power to handle big data. Minimum Error Entropy (MEE) principle is an important approach in Information Theoretical Learning (ITL). It is widely applied and studied in various fields for its robustness to noise. In this talk, we introduce a reproducing kernel-based distributed MEE algorithm, DMEE, which is designed to work with both fully supervised data and semi-supervised data. Among many strategies of distributed learning, the divide-and-conquer approach is employed, so there is no inter-node communication overhead. We show that DMEE significantly reduces the computational complexity and memory requirement on single computing nodes.
会议ID：299286771
主讲人简介：
胡婷，理学博士，武汉大学数学与统计学院副教授，主要从事机器学习，统计学习理论等方面的研究。主持国家自然科学基金青年项目和面上项目各一项，在 Applied and Computational Harmonic Analysis，Journal of Machine Learning Research， Inverse Problems，IEEE Transactions on Signal Processing， Constructive Approximation，Studies in Applied Mathematics等应用数学和机器学习领域中有影响力的期刊发表了一系列学术论文。
