讲座题目: Model-Based and Physics-Informed Deep Learning NN Structures: Application in Acoustic and Infrared Imaging

讲座时间：2024年11月27日（星期三）9:45

讲座地点：勤园12-504

讲座简介：

Neural Networks (NN) has been used in many area with great succes. When a NN’s structure (Model) is given, during the training steps, the parameters of the model is determined using an appropriate criterion and an optimization algoriothm (Training).Then, the trained model can be used for the prediction or inference step (Testing). As there are also many hyperparameters, related to the optimization criteria and optimization algorithms, a validation step is necessary before its final use.One of the great difficulties is the choice of the NN’s structure. Even if there are many ”on the shelf” networks, selecting or proposing a new appropriate network for a given data, signal or image processing, is still an open problem.In this presentation, we consider this problem using model based signal and image processing and inverse problems methods. We classify the methods in five classes, based on:

1- Explicite analytical solutions, 

2- Transform domain decomposition, 

3- Operator decomposition, 4- Unfolding the optimization algorithms, and 

5- Physics Informed NN methods (PINN).

Some examples of application in Acoustical and infrared imaging will be presented.

报告人介绍：

Prof Ali Djafari

Distinguished Professor at Paris-Saclay University (Univ. Paris 11) France, Research Director of French National Research Center (CNRS). Pioneer and chairman of International Conference on Maximum Entropy and Bayesian Approaches (50 years); Top talent of Zhejiang Invited foreign-experts; Laureate of the Westlake Prize of Zhejiang for significate contribution of foreign experts; Laureate of the International scientific cooperation of Zhejiang.

He received the master and two PhD degrees from University of Paris 11 respectively in 1980, 1994 and 1998 respectively. His proposed Gaussian Porter image segmentation algorithm, fast Bayesian variational method, hyperparameter Bayesian inference method, etc., have been highly recognized by the international academic and industrial communities, and have been widely applied in the fields of non-destructive detection, mechanical fault diagnosis, medical image recognition, and industrial big data analysis. His methods and inventions have been directly adopted by Airbus, Thales, Dassault, CEA, etc. He has presided 31 projects (with 10 million euros funding), published over 300 papers, 2 monographs and 12 textbooks; He supervised 21 doctoral and 31 master's students.

Dr. Chu Ning, professorate engineer, IEEE senior member, and vice-chairman of the Zhejiang Acoustic Society; Chief researcher with Zhejiang Shangfeng Special Blower Company Ltd., China. He received B.S. degree in information engineering from the National University of Defense Technology, Changsha, China in 2006, and the M.S. and Ph.D. in automatic, signal image processing from the Paris-Saclay University, France in 2010 and 2014, respectively, and the postdoc at EPFL Switzerland.

His research interests are acoustic source imaging, infrared detection and Bayesian inference in machine fault prognosis. He has invented “Industrial Lung System” for green ventilation equipment, reported by CCTV2 in 2022, selected into the list of Zhejiang industrial Internet platform, and best cases of China intelligent manufacturing. In recent 5 years, he published more than 24 top journal papers and own 31 China invention patents, presided 3 national and provincial research projects.