all InfoSec news
To Trust or Not To Trust Prediction Scores for Membership Inference Attacks. (arXiv:2111.09076v2 [cs.LG] UPDATED)
May 2, 2022, 1:20 a.m. | Dominik Hintersdorf, Lukas Struppek, Kristian Kersting
cs.CR updates on arXiv.org arxiv.org
Membership inference attacks (MIAs) aim to determine whether a specific
sample was used to train a predictive model. Knowing this may indeed lead to a
privacy breach. Most MIAs, however, make use of the model's prediction scores -
the probability of each output given some input - following the intuition that
the trained model tends to behave differently on its training data. We argue
that this is a fallacy for many modern deep network architectures.
Consequently, MIAs will miserably fail …
More from arxiv.org / cs.CR updates on arXiv.org
One-shot Empirical Privacy Estimation for Federated Learning
1 day, 4 hours ago |
arxiv.org
Transferability Ranking of Adversarial Examples
1 day, 4 hours ago |
arxiv.org
A survey on hardware-based malware detection approaches
1 day, 4 hours ago |
arxiv.org
Explainable Ponzi Schemes Detection on Ethereum
1 day, 4 hours ago |
arxiv.org
Jobs in InfoSec / Cybersecurity
SOC 2 Manager, Audit and Certification
@ Deloitte | US and CA Multiple Locations
Information Security Engineers
@ D. E. Shaw Research | New York City
Cyber Security Architect - SR
@ ERCOT | Taylor, TX
SOC Analyst
@ Wix | Tel Aviv, Israel
Associate Director, SIEM & Detection Engineering(remote)
@ Humana | Remote US
Senior DevSecOps Architect
@ Computacenter | Birmingham, GB, B37 7YS