Calibrating Noise to Sensitivity in Private Data Analysis¶
Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006). Calibrating Noise to Sensitivity in Private Data Analysis. Theory of Cryptography (TCC 2006), Lecture Notes in Computer Science 3876, Springer.
Cited by¶
4 citations across 4 artifacts.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Identifiability
- The privacy mechanism adds calibrated random noise so that the induced equivalence class — the set of databases that could have produced the observed output — is large and, crucially, contains both the database with the individual and the one without.
This sourceFounds differential privacy: calibrated noise makes neighboring databases (with and without an individual) nearly indistinguishable, enlarging the equivalence class.
Supported in partVerified against the publisher's abstract
The abstract states that privacy is protected by perturbing true answers with calibrated random noise and characterizes privacy as indistinguishability of transcripts, but never states the 'induced equivalence class' framing.
“To protect privacy, the true answer is perturbed by the addition of random noise generated according to a carefully chosen distribution, and this response, the true answer plus noise, is returned to the user.”
- The privacy mechanism adds calibrated random noise so that the induced equivalence class — the set of databases that could have produced the observed output — is large and, crucially, contains both the database with the individual and the one without.
Mechanisms¶
- Differencing Attack Scan
- This is the canonical reason privacy loss must be treated as cumulative rather than certified one query at a time.
This sourceEstablishes that a sequence of adaptive queries consumes a cumulative privacy budget.
- This is the canonical reason privacy loss must be treated as cumulative rather than certified one query at a time.
- Noise or Randomization Release
- The discipline is to fix the protection budget before the data is seen and let the noise follow from it
This sourceDerives the noise distribution from the chosen privacy parameter and the query's sensitivity.
- The discipline is to fix the protection budget before the data is seen and let the noise follow from it
- Privacy-Preserving Telemetry View
- Differential privacy is the framework for releasing aggregate statistics — telemetry included — with a bounded, quantified per-entity disclosure guarantee, and privacy-preserving telemetry systems apply it to metric collection.
This sourceDifferential privacy releases statistical information while bounding how much any one person's data can affect the output distribution.
- Differential privacy is the framework for releasing aggregate statistics — telemetry included — with a bounded, quantified per-entity disclosure guarantee, and privacy-preserving telemetry systems apply it to metric collection.
Verification¶
Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.
Does it back the claim? Read against the text for 1 of 4 citations: 1 supported in part. Each verdict is shown under its citation below, with what in the work backs the sentence.
Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.
See how references were verified.
Registry ID ref:f4b426b57e5e · see in the full table