Nicolas Ruiz · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.26460 · cited by 0 works (OpenAlex) · open-access PDF
Aus einer vorgeschlagenen alternativen Formulierung von randomisierten Matrizen, die mit differentiellen Datenschutzgarantien vereinbar sind, zeigen wir, dass unter der Randomisierung auf Attributebene und wo die Randomisierung der gesamten Verteilung durch Kronecker-Produkte gekennzeichnet ist, jeder Kreuzpunkt des anonymisierten Datensatzes mit unterschiedlichen Attributen dem ursprünglichen Kreuzmoment entspricht, der durch Abschwächungsparameter multipliziert wird, selbst durch den Teil des Gesamt-Datenschutzbudgets bestimmt, der jedem Attribut zugewiesen wird. Diese Charakterisierung ermöglicht explizite und transparente Kontrollen der multivariatenbeziehungen Erhaltungen in der differentialen Privatsphäre durch Allokation von Datenschutzbudgets. Eine einfache numerische Illustration zeigt,
Machine translation (Meta M2M100); the English original follows.
This paper derives an exact attenuation formula for multivariate cross-moments under differential privacy achieved through randomized response. From a proposed alternative formulation of randomization matrices compliant with differential privacy guarantees we show that, under attribute-level randomization and where the randomization of the whole distribution is characterized by Kronecker products, any cross-moment of the anonymized dataset involving distinct attributes is equal to the original cross-moment multiplied by attenuation parameters, themselves determined by the part of the overall privacy budget allocated to each attribute. This characterization enables explicit and transparent controls of multivariate relationships preservations in differential privacy through privacy budget allocation. A simple numerical illustration demonstrates that identical total budgets can yield substantially different covariance-preservation profiles. The results also make the multiplicative loss of higher-order interactions explicit and thus can support mechanism design guided by analytical priorities.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Differential privacy, Multivariate statistics, Multiplicative function, Randomized response, Kronecker product, Differential (mechanical device)
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26460, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange