Talks in EE: A Graduate Program at Tsinghua University, and Non-Gaussian Probability Laws in Deep Neural Networks, Jan. 22, at 14:00
You are cordially invited to the seminar, which will be given by Prof. Ercan E. Kuruoğlu, this Wednesday, January 22, 2025, at 14:00 hours.
At the beginning, there will be a talk on the Graduate Program of Tsinghua University Shenzhen International Graduate School, Institute of Data Science and Information.
Talk 1: “Tsinghua Shenzhen International Graduate School Institute of Data Science and Information (iDI) Graduate Program”
Talk 2: “Non-Gaussian Probability Laws in Deep Neural Networks”
Place: Electrical and Electronics Engineering Seminar Room
Date and Time: Wednesday, January 22, 2025 / 14:00 hours.
Talk 1:
Tsinghua Shenzhen International Graduate School
Institute of Data Science and Information (iDI)
Graduate Program
In this talk, I will present the Graduate Program of Tsinghua University Shenzhen International Graduate School, Institute of Data Science and Information. Tsinghua University appears at 25th position in global ranking according to QS Ranking, 12th according to Times World Universities Ranking, and 22th according to Shanghai Ranking. The Shenzhen campus of Tsinghua provides graduate education in English. A large percentage of students get full academic scholarship covering tuition and living expenses. This talk will provide ample information on study conditions and a question and answer session will follow.
Talk 2:
Non-Gaussian Probability Laws in Deep Neural Networks
The deep neural networks are envisioned as deterministic nonlinear mappings without memory. This assumption leads to lack of means for reliability/risk analysis and non-interpretability of NN systems. This requires the development of a probabilistic formulation of NNs. A common assumption among almost all work towards this direction is the Gaussian assumption of Neural Network variables. In this talk, we challenge this assumption and provide theoretical and experimental evidences for non-Gaussian laws in particular following Levy-stable characteristics. Next we discuss the problems of robustness and data compression.
Presenter: Prof Ercan E Kuruoglu
Ercan Engin Kuruoglu received M.Phil. and Ph.D. degrees in information engineering from the University of Cambridge, Cambridge, U.K., in 1995 and 1998, respectively. In 1998, he joined Xerox Research Center Europe, Cambridge. He was an ERCIM fellow in 2000 with INRIA-Sophia Antipolis, France. In January 2002, he joined ISTI-CNR, Pisa, Italy where he became a Chief Scientist in 2020. Currently, he is a Full Professor at Tsinghua-Berkeley Shenzhen Institute since March 2022. He served as an Associate Editor for the IEEE Transactions on Signal Processing and IEEE Transactions on Image Processing. He was the Editor in Chief of Digital Signal Processing: A Review Journal between 2011-2021. He is currently co-Editor-in-Chief of Journal of the Franklin Institute, responsible for Data Science and Signal Processing Section. He acted as a Technical co-Chair for EUSIPCO 2006 and a Tutorials co-Chair of ICASSP 2014. He is a member of the IEEE Technical Committee (TC) on Machine Learning for Signal Processing, andis Vice-Chair for IEEE TC on Image, Video and Multidimensional Signal Processing. He is also a member of the IEEE Data Collections and Challenges Committee. He was a plenary speaker at ISSPA 2010, IEEE SIU 2017, Entropy 2018, MIIS 2020, IET Radar 2023 and tutorial speaker at IEEE ICSPCC 2012. He was an Alexander von Humboldt Experienced Research Fellow in the Max Planck Institute for Molecular Genetics in 2013-2015. Prof. Kuruoglu’s research interests are in the areas of statistical signal and image processing, Bayesian machine learning and information theory with applications in remote sensing, environmental sciences, telecommunications and computational biology.