Title:An Introduction to Independent Component Analysis (ICA): methods, algorithms and applications
Time:13:00-17:30, December 23-24, 2019.
Location:Room 126, Fansun Building, Nankai University.
A fundamental problem in neural network research, and many other disciplines, isfinding a suitable representation of multivariate data. For computational simplicity, the representation is often sought as a linear transformation of the original data. Different from well-known linear transformation methods (such as PCA, factor analysis, and projection pursuit), the Independent Component Analysis (ICA) is a method aiming to represent the data via linearly combining hidden components which are “non-Gaussian” and “statistically independent”. Such a representation seems to capture the essential structure of the data in many applications, useful for feature extraction and signal separation. This short course introduces the basic theory, numerical algorithms, and recent research works on ICA, with applications to neuroscience, genetics and finance.
Prerequisite: Basic knowledge of statistical inference (例如,参数估计方法及统计性质); interest in statistical analysis of big data, and statistical learning methods.
Important and/or recent journal articles in the area will be discussed in lectures.
Chunming Zhang, Professor,
Department of Statistics,
University of Wisconsin-Madison, USA.