Speaker:

Zhi-Hua Zhou, National Key Laboratory for Novel Software Technology, Nanjing University

Title:

Long March of Margin Theory for AdaBoost

Location:

Owheo G34 - 1:00 pm, Friday 15 April

Abstract:

AdaBoost is a mainstream ensemble learning approach with high impact in machine learning and beyond. One particularly attractive property lies in the mysterious phenomenon that AdaBoost seems less prone to overfitting. There was a long thread theoretical study about this mystery. In this talk we will give an introduction and show some insight for algorithm design.

Bio:

Zhi-Hua Zhou is a Professor and Founding Director of the LAMDA Group at Nanjing University, China. His research interests are mainly in artificial intelligence, machine learning and data mining. He authored the book "Ensemble Methods: Foundations and Algorithms", and published more than 100 papers in top-tier journals and conference proceedings. His work received more than 20,000 citations, with a h-index of 71. He also holds many patents and has good experiences in industrial applications. He has received various awards, including the National Natural Science Award of China (premium science award in China), the IEEE CIS Outstanding Early Career Award, the Microsoft Professorship Award, 14 international journal/conference paper/competition awards, etc. He serves as the Executive Editor-in-Chief of "Frontiers of Computer Science", Associate Editor-in-Chief of "Science China", and Associate Editor of ACM TIST, IEEE TNNLS, etc. He founded ACML (Asian Conference on Machine Learning) and served as General Chair/Co-Chair of many conferences including ADMA'12, PAKDD'14 and ICDM'16, Program Committee Chair/Co-Chair of many conferences including SDM'13, ICDM'15 and IJCAI'15 Machine Learning Track, and Area Chair for NIPS, ICML, AAAI, IJCAI, KDD, ICDM, etc. He also serves as Advisory Committee member for IJCAI 2015-2016, and Steering Committee Member of PAKDD and PRICAI. He was elected to AAAI Fellow in 2016 for significant contributions to ensemble methods and learning from multi-labeled and partially-labeled data. He is also an Fellow of the IEEE, IAPR, IET/IEE, CCF, and an ACM Distinguished Scientist.