About the event
Large-scale distributed collection of contextual information is often essential in order to gather statistics and train machine learning models. The ability to do so in a privacy-preserving way enables a number of computational scenarios that would be hard, or outright impossible, to realize without strong security guarantees. In this talk, we present the design and deployment of practical techniques for privately gathering statistics from large data streams and sharing information in a secure way. We present and build on efficient cryptographic protocols for private data sharing and private data aggregation and efficient data structures; then, we show how to use these techniques to instantiate real-world privacy-friendly modeling systems. If time permits, we will then focus on how to identify and quantify possible privacy leakage from privately shared and aggregated data, as well as from models trained with private data.
Emiliano De Cristofaro (UCL, UK) is a Reader (Associate Professor) in Security and Privacy Enhancing Technologies at University College London (UCL). Prior to joining UCL in 2013, he spent two years as a Research Scientist at Xerox PARC in Palo Alto. In 2011, he received a PhD in Networked Systems from the University of California Irvine, advised - mostly while running on the beach - by Gene Tsudik. His research interests include privacy technologies, systems security, as well as Internet and Web measurements. He co-chaired the Privacy Enhancing Technologies Symposium (PETS) in 2013-2014 and the security track at WWW 2018. His relatively up-to-date homepage is at https://emilianodc.com