[SEMINAR] July 26th at 4pm CEST - Data Synthesis with Differential Privacy

On July 26th at 4pm CEST, prof. Chia-Mu Yu from National Yang Ming Chiao Tung University  will deliver a talk on "Data Synthesis with Differential Privacy".

Please find below abstract and bio.

The event will be in person in Room 2AB/40 in Torre di Archimede, but you can also join the event on zoom: 

<https://unipd.zoom.us/j/81497268111?pwd=ZXk3VnprVStyT1NHa0xmckhPbVhhUT09>

Meeting ID: 814 9726 8111
Passcode: 952331

You are all warmly invited to attend!
Mauro Conti

*Abstract*
Despite increased demand for valuable data, the privacy concerns associated with sensitive datasets present a barrier to data sharing. One may use generative models to generate synthetic data, but unfortunately, the synthetic data in this way may still contain sensitive information. Currently, differential privacy is the golden standard of data privacy. Thus, an alternative approach is to construct differentially private generative models that quantitatively ensure the disconnection between the sensitive data and synthetic data. In this talk, we will go through the idea of differential privacy and briefly describe the recent research progress on the development of differentially private generative models.

*Bio*
Chia-Mu Yu is currently an associate professor at National Yang Ming Chiao Tung University, Taiwan. He had academic visits at IBM Thomas J. Watson research center, Harvard University, Imperial College London, University of Padova, and the University of Illinois at Chicago. He Received Hwa Tse Roger Liang Junior Chair Professor, MOST Yong Scholar Fellowship, ACM/IICM K. T. Li Young Researcher Award, Observational Research Scholarship from Pan Wen Yuan Foundation, and MOST Project for Excellent Junior Research Investigators, Taiwan. His research interests include differentially private mechanism design, cloud storage security, and IoT security.