数学与系统科学学院学术报告预告
报告一
时间：2019年7月2日 星期二 9:00-9:40
地点：第一教学楼1211室
题目：On the Restricted Boltzmann Machines for Deep Learning
报告人简介： 杨力华，中山大学数学学院教授、博士生导师。先后在湖南师范大学, 北京师范大学和中山大学获得学士, 硕士和博士学位。1996年至1998年在中科院数学研究所从事博士后工作。历任全国计算数学学会理事、广东省计算数学学会理事长、IEEE高级会员。研究领域为: 函数逼近与小波分析, 信号处理与机器学习， 迄今为止发表论文100余篇，合作出版专著一本，译著三本，教材一本。
内容提要：
Based on the structure of Deep Belief Networks, we discuss the mathematical problems in deep learning, including the expression power and learning algorithm of Boltzmann Machines。
报告二
时间：2019年7月2日 星期二 9:45-10:25
地点：第一教学楼1211室
题目：Tensor Decomposition for Multilayer Networks Clustering
报告人简介：陈川，现任中山大学数据科学与计算机学院副研究员。2016年于香港浸会大学数学系获得博士学位，2016-2017 年于比利时鲁汶大学电子工程系任博士后研究员。主要研究方向为：数值优化，机器学习及大数据分析。近年来发表SCI索引国际期刊论文及AAAI, ICML, IJCAI等国际会议论文近30篇。担任 IEEE TIP/TSP等多份国际期刊审稿人，担任IJCAI ECAI CBPM等多个国际学术会议的程序委员会成员及论坛主义，广东省计算机协会区块链专委会委员。现主持国家青年科学基金，广东省创新研究基金，CCF协会项目基金, 与包括美图，微信，华为，中国电信，平安保险在内多家企业开展横向项目研究。
内容提要：
Clustering on multilayer networks has been shown to be a promising approach to enhance accuracy. Various multilayer networks clustering algorithms assume all networks derive from a latent clustering structure and jointly learn the compatible and complementary information from different networks to excavate one shared underlying structure. However, such assumptions are in conflict with many emerging real-life applications due to the existence of noisy/irrelevant
networks. A key challenge here is to integrate different data representations automatically to achieve better predictive performance. To address this issue, we propose Centroid-based Multilayer Network Clustering (CMNC), a novel approach which can divide irrelevant relationships into different network groups and uncover the cluster structure in each group simultaneously. The multilayer networks are represented within a unified tensor framework for simultaneously capturing multiple types of relationships between a set of entities. By imposing the rank-(Lr; Lr; 1) block term decomposition with nonnegativity constraints, we are able to have well interpretations on the multiple clustering results based on graph cut theory. Numerically, we transform this tensor decomposition problem to an unconstrained optimization, thus can solve it efficiently under the nonlinear least squares (NLS) framework. Extensive experimental results on synthetic and real-world datasets show the effectiveness and robustness of our method against noise and irrelevant data.
报告三
时间：2019年7月2日 星期二 10:30-11:10
地点：第一教学楼1211室
题目：Gersgorin-type E-eigenvalue localization sets and the positive definiteness of even order tensors
报告人简介：赵建兴（Jianxing Zhao），男，1981年1月生，贵州民族大学数据科学与信息工程学院,教授，应用数学博士，主要从事数值代数、特殊矩阵（张量）特征值估计等方向的研究，主持国家自然科学基金青年基金项目、贵州省科技厅科学技术基金、贵州省教育厅科技拔尖人才支持项目各一项，入选贵州省普通高等学校科技拔尖人才（2016年），中组部“西部之光”访问学者（2018年），以第一作者或通讯作者发表SCI期刊论文20余篇。
内容提要：
In this talk, some existing Z-eigenvalues of tensors are first recalled. Secondly, a Gersgorin-type E-eigenvalue localization set with applications to judge the positive (semi-)defniteness of tensors is introduced. As an application, an upper bound for the Z-spectral radius of weakly symmetric nonnegative tensors is obtained. Finally, numerical examples are given to verify the theoretical results.
欢迎老师、同学们参加！
数学与系统科学学院
2019年6月26日
