Invited Talks

Invited Talk 1:A Multi-objective Learning Model for Deep Neural Networks

Prof. Maoguo Gong

Speaker : Prof. Maoguo Gong
Abstract :  Hierarchical deep neural networks are currently popular learning models for imitating the hierarchical architecture of human brain. Single-layer feature extractors are the bricks to build deep networks. Sparse feature learning models are popular models that can learn useful representations. But most of those models need a user-defined constant to control the sparsity of representations. In this report, we introduce a multiobjective sparse feature learning model based on the autoencoder. The parameters of the model are learnt by optimizing two objectives, reconstruction error and the sparsity of hidden units simultaneously to find a reasonable compromise between them automatically. We design a multiobjective induced learning procedure for this model based on a multiobjective evolutionary algorithm. In the experiments, we demonstrate that the learning procedure is effective, and the proposed multi-objective model can learn useful sparse features.
Biography :  Dr. Maoguo Gong received the B. Eng degree and Ph.D. degree from Xidian University. Since 2006, he has been a teacher of Xidian University. He was promoted to associate professor and full professor in 2008 and 2010, respectively, both with exceptive admission. Gong’s research interests are broadly in the area of computational intelligence, with applications to optimization, learning, data mining and image understanding. He has published over one hundred papers in journals and conferences, and holds fifteen granted patents as the first inventor. He is leading or has completed over ten projects as the PI, funded by the National Natural Science Foundation of China, the National High Technology Research and Development Program (863 Program) of China and others. He was the recipient of the prestigious National Program for Support of Top-notch Young Professionals (selected by the Central Organization Department of China), the Excellent Young Scientist Foundation (selected by the National Natural Science Foundation of China), the New Century Excellent Talent in University (selected by the Ministry of Education of China), the Young Teacher Award by the Fok Ying Tung Education Foundation, the Young Scientist Award of Shaanxi Province, the New Scientific and Technological Star of Shaanxi Province, the Elsevier SCOPUS Young Researcher Award of China, and the National Natural Science Award of China. He is the Vice-Chair of IEEE CIS Task Force on Memetic Computing, executive committee member of Chinese Association for Artificial Intelligence, senior member of IEEE and Chinese Computer Federation, Editorial Board member for five journals including the International Journal of Bio-Inspired Computation, reviewer for over ten journals including IEEE Transactions on Evolutionary Computation, IEEE Transactions on Geoscience and Remote Sensing, and European Journal of Operational Research.

Invited Talk 2:Label Noise Robust Boosting Algorithm Based on a Non-convex Loss Function and the Numerically Stable Base learners

Prof. Qiguang Miao

Speaker : Prof. Qiguang Miao
Abstract :  AdaBoost has attracted much attention in the machine learning community because of its excellent performance in combining weak classifiers into strong classifiers. However, AdaBoost tends to over-fit to the noisy data in many applications. Accordingly, improving the anti-noise ability of AdaBoost plays an important role in many applications. The sensitiveness to noisy data of AdaBoost stems from the exponential loss function which puts unrestricted penalties to misclassified samples with very large margins. In this paper, we propose two Boosting algorithms, referred to as RBoost1 and RBoost2, which are more robust to noisy data with comparison to AdaBoost. RBoost1 and RBoost2 optimize a non-convex loss function of the classification margin. Because the penalties to misclassified samples are restricted to an amount less than one, RBoost1 and RBoost2 do not over-focus on samples that are always misclassified by the previous base learners. Besides the loss function, at each Boosting iteration, RBoost1 and RBoost2 use numerically stable ways to compute the base learners. These two improvements both contribute to the robustness of the proposed algorithms to the noisy training and testing samples. Experimental results on the synthetic Gaussian dataset, the UCI datasets and a real malware behavior dataset illustrate that the proposed RBoost1 and RBoost2 perform better when training datasets contain noisy data.
Biography :  Since 2006, Dr. Qiguang Miao has been a Full Professor in School of Computer Science and Technology at Xidian University, Xi'an, China, and the Director of Machine Learning and Intelligent Image Processing laboratory (MLIP). His research interests include pattern recognition, machine learning, and malware behavior analysis. In 2012, Dr. Miao was selected as a member of the program for New Century Excellent Talents in University of China by the Ministry of Education (MOE). In 2012, he obtained the third prize of the Chongqing Science and Technology Award. So far, Dr. Miao is a Member of IEEE; Member of ACM ; IEEE CS Society; Senior Member of China Computer Federation (CCF) and AC of CCF YOCSEF , an Executive member of Artificial Intelligent and pattern Recognition Council of CCF, and Members of Editorial board for IOT, etc.

Copyright@SDSBC 2015