Security and Privacy in Machine Learning: Threat Models and Mitigation Measures
TECoSA Seminar with Dr Raksha Ramakrishna
Time: Thu 2022-09-01 15.00 - 16.00
Video link: https://kth-se.zoom.us/j/66857695267
ABSTRACT: Machine Learning (ML) is becoming ubiquitous by the day, leading to an increase in the privacy attacks that are possible thereby risking the security of both the data used for training the ML models and the learned models themselves which could be considered intellectual property. In this seminar, we will discuss the taxonomy of threat models for ML models and highlight examples of privacy attacks and possible mitigation measures for them. In particular, property inference attacks in the context of centralized and federated ML models will also be discussed based on our recent work.
BIO: I received my Ph.D degree in Electrical Engineering from Arizona State University (ASU) in 2020 where I worked in the SINE lab directed byProf. Anna Scaglione . I obtained an MS degree from ASU in 2017 and BE degree in Electronics and Communications Engineering from Rashtreeya Vidyalaya College of Engineering, Bangalore, India in 2014. I am currently a Postdoctoral Researcher at the Division of Network and System Technology with Prof. György Dán’s group, and also affiliated to the Center for Trustworthy Edge Computing Systems and Applications (TECoSA) .
TECoSA hosts a guest speaker on the first Thursday of each month during term time. The seminars are free and all are welcome to attend. This Autumn, all seminars will be available via Zoom (with some also IRL at KTH Campus). Please see the TECoSA Seminar Series homepage for details! www.tecosa.center.kth.se/tecosa-seminar-series/