all InfoSec news
Reconstructing Training Data with Informed Adversaries. (arXiv:2201.04845v1 [cs.CR])
Jan. 14, 2022, 2:20 a.m. | Borja Balle, Giovanni Cherubin, Jamie Hayes
cs.CR updates on arXiv.org arxiv.org
Given access to a machine learning model, can an adversary reconstruct the
model's training data? This work studies this question from the lens of a
powerful informed adversary who knows all the training data points except one.
By instantiating concrete attacks, we show it is feasible to reconstruct the
remaining data point in this stringent threat model. For convex models (e.g.
logistic regression), reconstruction attacks are simple and can be derived in
closed-form. For more general models (e.g. neural networks), …
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
SOC 2 Manager, Audit and Certification
@ Deloitte | US and CA Multiple Locations
EY GDS Internship Program - SAP, Cyber, IT Consultant or Finance Talents with German language
@ EY | Wrocław, DS, PL, 50-086
Security Architect - 100% Remote (REF1604S)
@ Citizant | Chantilly, VA, United States
Network Security Engineer - Firewall admin (f/m/d)
@ Deutsche Börse | Prague, CZ
Junior Cyber Solutions Consultant
@ Dionach | Glasgow, Scotland, United Kingdom
Senior Software Engineer (Cryptography), Bitkey
@ Block | New York City, United States