'''
讲座人：Chunhua Shen (A tenured senior lecturer at School of Computer Science, The University of Adelaide, Australia)讲座时间：2013.4.23（周二）上午 10:00-11:00讲座地点：复旦大学张江校区计算机楼 405联系人：路红 honglu@fudan.edu.cnAbstract: Consideration of the primal and dual problems together leads to important new insights into the characteristics of boosting algorithms. Based on this insight, a general boosting framework is then proposed, which can be used to design new boosting algorithms. A wide variety of machine learning problems essentially minimize a regularized risk function. We show that the proposed boosting framework can accommodate various loss functions and different regularizers in a fully-corrective optimization fashion. A large body of fully-corrective boosting algorithms can actually be efficiently solved and no sophisticated convex optimization solvers are needed. We also demonstrate that some conventional boosting algorithms like AdaBoost can be interpreted in our framework---even though their optimization is not fully corrective. An extension of this framework can also be used to design new multi-class boosting or more general boosting methods for structured output learning.Bio: Chunhua Shen is a tenured senior lecturer at School of Computer Science, The University of Adelaide. Before that he worked as a researcher at the computer vision program at NICTA (National ICT Australia), Canberra Research Laboratory for about six years. His research interests are in the intersection of computer vision and statistical machine learning. Recent work has been on real-time object detection, large-scale image retrieval and classification, and scalable nonlinear optimization. He received BSc (at dept. of intensive instruction), MSc (at dept. electronic engineering) both from Nanjing University, PhD from The University of Adelaide. He held an adjunct researcher position at College of Engineering & Computer Science, Australian National University from 2006 to 2011. From 2012 to 2016, he holds an Australian Research Council Future Fellowship.
'''
