Maria Rigaki and Sebastián García wrote a new blog post mapping how much information ML systems leak. She reviewed the recent techniques, important attacks and why they're possible in the first place. Read the text titled Machine Learning Leaks and Where to Find Them.
We invite you to the CyberSec&AI Connected2020 security conference, which will take place on October 8 in Prague, London, New York and San Francisco. The event organized by CTU and our partner Avast will be streamed online for participants all over the world. The conference also includes technical workshops with Call for Presentations (June 30).
The research paper "On Model Evaluation under Non-constant Class Imbalance" by Jan Brabec and Tomáš Komárek got accepted to the A-rate conferenceICCS 2020. Great success!
