Welcome to FLUIDOS TechTalk #2: ‘Data forging is harder than you think’
FLUIDOS continues to enrich its TechTalks series with a thought-provoking presentation by Mohamed Suliman on the topic of data forging. The paper “Data forging is harder than you think” was recently showcased at the “Privacy Regulation and Protection in Machine Learning” workshop during the International Conference on Learning Representations (ICLR 2024).
The session delved into the complexities of data forging attacks within machine learning, focusing on privacy and security implications. In this paper, Mohamed Suliman, alongside co-authors Swanand Kadhe, Anisa Halimi, Douglas Leith, Nathalie Baracaldo and Ambrish Rawat, explore the challenges of crafting data batches that mimic original training data without actual learning, posing significant privacy risks.
This webinar series aims to share cutting-edge research and foster collaboration among project partners and industry experts. Don’t forget to subscribe to our YouTube channel for full video presentations and stay updated with forthcoming sessions in this enlightening series!It was presented during Eurosys 2024, at the 1st International Workshop on MetaOS for the Cloud-Edge-IoT Continuum (MECC 2024) that took place in April.
Don’t forget to subscribe to our YouTube channel to stay up to date with all the latest FLUIDOS advancements and forthcoming TechTalks!
