报告题目: Non-Gaussian Statistical Models---Theory, Methods, and Applications

报告人：马占宇副教授

报告时间：2017年1月12日(周四) 16:20-17:30

报告地点：计算机学院8号楼208会议室

摘要：

Recent research demonstrate that the usage of non-Gaussian statistical models is advantageous in applications where the data are not Gaussian distributed. With conventionally applied model estimation methods, e.g., maximum likelihood estimation and Bayesian estimation, we cannot derive analytically tractable solution for non-Gaussian statistical models. In order to obtain closed-form solution, we extend the commonly used variational inference (VI) framework via lower-bound approximation, by utilizing convexity/relative convexity of the integrands in the non-Gaussian distributions. In this presentation, we introduce the principles of the extended variational inference (EVI) and demonstrate its advantages in non-Gaussian mixture models and bounded support matrix factorization. In addition to this, we also introduce our recent work about non-Gaussian neutral vector analysis, which proposed a novel framework for neutral vector decorrelation based on the complete neutrality, the exchangeability, and the permutation properties. The advantages of non-Gaussian statistical models are demonstrated in real life applications, such as speech coding, 3D depth map enhancement, DNA methylation analysis, EEG signal classification.

 

个人简介

马占宇，北京邮电大学副教授，丹麦奥尔堡大学兼职副教授，北邮信通院-邦赢彩票大数据联合实验室主任，中国计算机学会计算机视觉专委会委员、副秘书长。曾于瑞典皇家理工学院攻读博士学位，并从事博士后研究。主要研究方向是模式识别与机器学习及其在非高斯概率模型、城市大数据建模、多媒体信号处理、生物医学信号处理、生物信息学等领域的应用。共在包括IEEE TPAMI、IEEE TNNLS、ICASSP、ECCV在内的期刊和会议上发表论文50多篇，担任NEUROCOMPUTING特刊主编和多个期刊的审稿人，获授权发明专利5项，相关技术被应用于多个实际系统中；主持包括国家自然科学基金、北京市自然科学基金在内的多个项目，入选2017年度北京市科技新星计划。