Learning and Identifying
Speaker:Prof. Luoqing Li
Faculty of Mathematics and Computer Science
Hubei University
Date & Time:30 Oct 2012 (Tuesday) 10:30
Organized by:Department of Mathematics


Learning theory and system identification are the art and science of building mathematical models of dynamic systems from observed input-output data. It can be seen as the interface between the real world of applications and the mathematical world of control theory and model abstractions. The core of estimating models is statistical theory. This talk will report the relationships between learning theory and identifying dynamical systems from noisy observations. A sparse algorithm for linear system identification from noisy measurements is explored. We are sure that the system identification community would benefit from an influx of new ideas from statistical learning theory.


Prof. Luoqing Li is currently the Professor of Hubei University. He obtained his Ph.D. in Beijing University. Prof. Li is interested in Harmonic Analysis and Approximation Theory, Wavelets, Learning Theory, Time-Frequency Analysis, Pattern Recognition and Image Processing. His recent publication include, Error analysis of stochastic gradient descent ranking, Learning rates of multi-kernel regression by orthogonal greedy algorithm, Hierarchical Feature Extraction with Local Neural Response for Image Recognition IEEE Transactions on Systems and Error Analysis for Matrix Elastic-Net Regularization Algorithms Similarity learning for object recognition based on derived kernel.