专题侠客群论剑-20150709

【15-21期VALSE Webinar活动】

报告嘉宾1俞扬(南京大学)
主持人孟德宇(西安交通大学)
报告时间:2015年7月9日周四晚20:00(北京时间)
报告题目:演化学习研究进展 [Slides]
文章信息
[1] Yang Yu, Xin Yao, and Zhi-Hua Zhou. On the approximation ability of evolutionary optimization with application to minimum set cover. Artificial Intelligence, 2012, 180-181:20-33.
[2] Chao Qian, Yang Yu and Zhi-Hua Zhou. Pareto ensemble pruning. In: Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI’15), Austin, TX, 2015, pp.2935-2941.
[3] Chao Qian, Yang Yu and Zhi-Hua Zhou. On constrained boolean Pareto optimization. In: Proceedings of the 23rd International Joint Conference on Artificial Intelligence (IJCAI’15), Buenos Aires, Argentina, 2015.
报告摘要:机器学习任务中往往包含非凸优化问题,以往基于凸放松的解决方案常带来很多局限,非凸优化问题的有效解决将为机器学习带来更大的发展空间。演化优化算法在许多复杂工业优化中已显示出卓越的优化性能,为解决非凸机器学习问题提供了选择,然而以往演化优化算法由于理论基础薄弱,难以深入应用于机器学习任务中。本次报告将汇报我们对于演化优化算法近似逼近能力的理论分析工作,该工作揭示了演化算法可以是目前最好的近似优化算法,并汇报基于该理论,将演化优化应用在机器学习任务中的一些结果。
报告人简介:俞扬,博士,南京大学计算机系副教授。主要研究领域为人工智能、机器学习、演化计算、数据挖掘。分别于2004年和2011年获得南京大学计算机科学与技术系学士学位和博士学位。2011年8月加入南京大学计算机科学与技术系、机器学习与数据挖掘研究所(LAMDA)从事教学与科研工作。曾获2013年全国优秀博士学位论文奖、2011年中国计算机学会优秀博士学位论文奖。发表论文20余篇,包括多篇Artificial Intelligence、JAIR、IEEE TEC、IJCAI、AAAI、KDD等国际一流期刊和会议,获得KDD’12 Best Poster、GECCO’11 Best Theory Paper、PAKDD’08 Best Paper、PAKDD’06数据挖掘竞赛冠军等论文和竞赛奖。任《Frontiers of Computer Science》青年副编辑、人工智能领域国际顶级会议IJCAI’15高级程序委员、IEEE计算智能协会演化计算技术委员会委员、IEEE计算智能协会数据挖掘与大数据分析技术委员会委员。

报告嘉宾2:周爱民 (华东师范大学)
主持人蓝振忠(CMU)
报告时间:2015年7月9日周四晚21:00(北京时间)
报告题目:Learning Guided Multiobjective Optimization [Slides]
文章信息
[1] H. Zhang, A. Zhou, S. Song, Q. Zhang, X. Gao, and J. Zhang, A self-organizing multiobjective evolutionary algorithm, 2015 (submit).
[2] A. Zhou, J. Sun, and Q. Zhang, An estimation of distribution algorithm with cheap and expensive local search, IEEE TEVC, 2015. (accepted)
[3] A. Zhou, and Q. Zhang, Are all the subproblems equally important? Resource allocation in decomposition based multiobjective evolutionary algorithms, IEEE TEVC, 2015. (accepted)
[4] W. Gong, A. Zhou, and Z. Cai, A multi-operator search strategy based on cheap surrogate models for evolutionary optimization, IEEE TEVC, 2015. (accepted)
[5] Z. Wang, Q. Zhang, A. Zhou, M. Gong, and L. Jiao, Adaptive replacement strategies for MOEA/D, IEEE TCYB, 2015. (accepted)
[6] A. Zhou, Y. Jin, and Q. Zhang, A population prediction strategy for evolutionary dynamic multiobjective optimization, IEEE TCYB, 44(1): 40-53,2014.
[7] A. Zhou, Q. Zhang, and Y. Jin, Approximating the set of Pareto-optimal solutions in both the decision and objective spaces by an estimation of distribution algorithm, IEEE TEVC, 13(5): 1167-1189, 2009.
[8] Q. Zhang, A. Zhou, and Y. Jin, RM-MEDA: a regularity model-based multiobjective estimation of distribution algorithm, IEEE TEVC, 12(1): 797-799, 2008.
报告摘要:The population of an evolutionary algorithm can be regarded as a data set that contains some kind of patterns. Although some evolutionary algorithms, such as estimation of distribution algorithms which utilize the probability graphic models to extract the patterns and surrogate assisted evolutionary algorithms which use regression methods, try to find the patterns and to guide the evolution process, there is still lack of a systematic work on using statistical and machine learning techniques to guide the evolutionary optimization. Furthermore, the area of machine and statistical learning contains a large broad of techniques but only a few ones have been utilized in the community of evolutionary computation. This talk tries to build a bridge from statistical and machine learning to evolutionary optimization especially evolutionary multiobjective optimization. The talk will cover the background on multi objective optimization, an example on how to use self-organizing maps to assist the search, a short survey on our recent work on this topic, and some conclusions and remarks for future work.
报告人简介:Aimin Zhou is currently an Associate Professor with the Shanghai Key Laboratory of Multidimensional Information Processing, and the Department of Computer Science and Technology, East China Normal University, Shanghai, China. He received the B.Sc. and M.Sc. degrees from Wuhan University, Wuhan, China, in 2001 and 2003, respectively, and the Ph.D. degree from University of Essex, Colchester, U.K., in 2009, all in computer science. His research interests include evolutionary computation and optimization, machine learning, image processing, and their applications. He has published over 40 peer-reviewed papers. He received the best paper award in IES 2014. He is an Associate Editor of the Swarm and Evolutionary Computation.

(Visited 262 times, 1 visits today)