讲座名称：面向智能信息处理的协同式神经动力学优化
讲座人：王钧 教授
讲座时间：4月10日9:00-11:00
讲座地点：长安校区会议中心101报告厅
讲座人介绍：
王钧，香港城市大学讲座教授。曾在大连理工大学、凯斯西储大学、北达科他大学和香港中文大学担任过多个学术职位，曾在美国空军阿姆斯特朗实验室、日本理化学研究所脑科学研究所、上海交通大学和华中科技大学担任多个短期访问职务。王钧教授在大连理工大学获得学士和硕士学位，并于凯斯西储大学获得博士学位。现在担任IEEE Transactions on Artificial Intelligence主编，曾任IEEE Transactions on Cybernetics主编。是IEEE Life Fellow、IAPR Fellow、香港工程科学院院士和欧洲科学院（Academia Europaea）外籍院士。是亚太神经网络（APNNA）杰出成就奖、IEEE CIS神经网络先驱奖、CAAI吴文俊人工智能科技成就奖、IEEE SMCS诺伯特-维纳奖等多项荣誉的获得者。
Jun Wang is a chair professor at City University of Hong Kong. Prior to it, he held various academic positions at Dalian University of Technology, Case Western Reserve University, University of North Dakota, and the Chinese University of Hong Kong. He also held various short-term visiting positions at USAF Armstrong Laboratory, RIKEN Brain Science Institute, Shanghai Jiao Tong University, and Huazhong University of Science and Technology. He received a B.S. and an M.S. from Dalian University of Technology and a Ph.D. degree from Case Western Reserve University. He is the Editor-in-Chief of the IEEE Transactions on Artificial Intelligence and was the Editor-in-Chief of the IEEE Transactions on Cybernetics. He is an IEEE Life Fellow, IAPR Fellow, and HKAE Fellow, and a foreign member of Academia Europaea. He is a recipient of the APNNA Outstanding Achievement Award, IEEE CIS Neural Networks Pioneer Award, CAAI Wu Wenjun AI Achievement Award, and IEEE SMCS Norbert Wiener Award, among other distinctions.
讲座内容：
过去四十年见证了神经动力学优化的诞生和发展，由于其生物学合理性和并行分布式信息处理的固有性质，神经动力学优化具有解决约束优化问题的强大潜力。近年来，面对全局优化、组合优化以及混合整数等挑战性优化问题的求解，协同式神经动力学优化的混合智能框架应运而生。在协同神经动力学优化框架中，多个初始状态多样化的神经动力学优化模型被并行用于分散局部搜索，并使用元启发式规则在局部收敛时重新定位神经元状态以避开局部最小值并走向全局最优解。该方法在理论上证明了其全局收敛性，同时也通过实验验证了有效性。在本次报告中，将从数据处理、特征选择、监督学习和压缩感知等方面开展介绍。
The past four decades have witnessed the emergence and growth of neurodynamic optimization, which has become a potentially powerful problem-solving tool for constrained optimization due to its inherent biological plausibility and parallel, distributed information-processing capabilities. In recent years, a hybrid intelligence framework known as collaborative neurodynamic optimization has been proposed to solve challenging optimization problems, including global, combinatorial, and mixed-integer optimization. In the context of collaborative neurodynamic optimization, multiple recurrent neural networks are employed to perform a scattered search for optimal solutions from different initial states, and a metaheuristic rule is used to reinitialize neuronal states, thereby repositioning the search for global optimal solutions. This approach has been theoretically proved to be globally convergent and experimentally demonstrated to be effective for many applications. In this talk, I will present several specific paradigms in this framework for data processing, feature selection, supervised learning, and compressive sensing.
主办单位：研究生院、人工智能学院、物理学院
活动报名二维码：
