報告題目:Sparse Machine Learning in a Banach Space
報告人:許躍生
報告時間:2020年10月25日(星期六)8:30—9:30
報告地點:線上講座(ZOOM ID: 935 1049 4677)
主辦單位:數學與數量經濟學院
主講人簡介:
許躍生,現任美國奧多明尼昂大學教授。本科、碩士畢業于中山大學,1989年在美國奧多明尼昂大學獲得博士學位。曾任美國西弗吉尼亞大學Eberly講席教授,雪城大學終身職正教授,中山大學國華講席教授。現擔任多個學術期刊編委。在劍橋大學出版社出版合著一部。論文發表在Applied and Harmonic Computational Analysis, SIAM Journal on Numerical Analysis, Inverse Problems, SIAM Journal on Imaging Science, IEEE Transactions on Medical Imaging等國際著名期刊,共計一百八十余篇。在計算數學、應用數學和基礎數學的多個領域都做出過重要學術貢獻。
報告內容簡介:
We will report in this talk recent development of kernel based machine learning. We will first review a basic classical problem in machine learning - classification, from which we introduce kernel based machine learning methods. We will consider two fundamental problems in kernel based machine learning: representer theorems and kernel universality. We will then elaborate recent exciting advances in sparse learning. In particular, we will discuss the notion of reproducing kernel Banach spaces and learning in a Banach space.