Skip to content
NewA calmer, faster ZeroHour
Researchinfo

BetweenCut: Private Heavy-Node Classification with Doubly Logarithmic Error in Tree Height

arXiv cs.CRby Ergute Bao ()1 min read1
Read at arxiv.org
AI summary · grok-4.7Summaries by Vani AI

BetweenCut privately finds heavy tree nodes with doubly logarithmic error in tree height under differential privacy.

The paper presents BetweenCut, an (ε,δ)-differentially private algorithm for finding heavy nodes in a tree. Prior methods incur additive error of Ω(log h) or Ω(√log h) because each record affects an entire root-to-leaf path. BetweenCut achieves O(log log h) additive error simultaneously for all nodes, independent of database size.

Key facts
  • Record-level privacy cost accumulates along each root-to-leaf path.
  • Prior additive error scales as log h or square-root log h.
  • BetweenCut reduces the margin to O(log log h) for all nodes.
  • The error bound does not depend on database size.
Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat
Tags
differential-privacytreesheavy-nodesalgorithmsprivacy

From the original article

1 min read at arxiv.org

Finding heavy nodes in a tree---those whose counts exceed a given threshold---is a building block for analysis and learning over structured data. Achieving record-level differential privacy (DP) without sacrificing accuracy is challenging because each record contributes to counts along an entire root-to-leaf path, allowing privacy costs to accumulate across levels.

arXiv cs.CR

By Ergute Bao · arxiv.org

Read the full article at arXiv cs.CR

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.