all InfoSec news
Concentrated Differential Privacy for Bandits
Feb. 16, 2024, 5:10 a.m. | Achraf Azize, Debabrota Basu
cs.CR updates on arXiv.org arxiv.org
Abstract: Bandits serve as the theoretical foundation of sequential learning and an algorithmic foundation of modern recommender systems. However, recommender systems often rely on user-sensitive data, making privacy a critical concern. This paper contributes to the understanding of Differential Privacy (DP) in bandits with a trusted centralised decision-maker, and especially the implications of ensuring zero Concentrated Differential Privacy (zCDP). First, we formalise and compare different adaptations of DP to bandits, depending on the considered input and …
arxiv critical cs.cr cs.it cs.lg data decision differential privacy foundation making math.it math.st privacy recommender systems sensitive sensitive data stat.ml stat.th systems understanding
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
G230ISSO2 - Mid Level Information System Security Officer - Cleared
@ NiSUS Technologies | Annapolis Junction, Maryland, United States
Security Incident Response Engineer
@ Oracle | JALISCO, Mexico
Security Compliance Specialist
@ Cloudflare, Inc. | Hybrid or Remote
Senior Security DevOps
@ SAP | Sofia, BG, 1407
Senior Cyber Security Engineer
@ Node.Digital | Dulles, Virginia, United States
Manager, Data Insights and Forensics
@ Kroll | New York City, United States