cd /news/ai-safety/revisiting-the-provable-auditable-pr… · home topics ai-safety article
[ARTICLE · art-117337] src=arxiv.org ↗ pub= topic=ai-safety verified=true sentiment=· neutral

Revisiting the Provable-Auditable Privacy Gap of DP-SGD

A new arXiv paper (2608.28934v1) proposes a lightweight defense framework that improves the empirical privacy lower bound of DP-SGD without increasing its theoretical privacy cost, addressing the gap between theoretical and audited privacy guarantees. The authors evaluate the framework across multiple audit constructions, models, and datasets, showing significantly improved empirical privacy on standard benchmarks.

read1 min views1 publishedSep 1, 2026

arXiv:2608.28934v1 Announce Type: new Abstract: Differential privacy (DP) has traditionally been used to provide theoretical upper bounds on an algorithm's stability to changing its training data. In modern private machine learning applications, achieving strong tradeoffs between utility and theoretical privacy is challenging, and thus one may optimistically hope that existing theoretical privacy analyses are loose. Recent work on privacy auditing has adopted a dual viewpoint, instead lower bounding the true privacy of an algorithm by constructing empirical distinguishing events. The auditing literature has thus far yielded a pessimistic outlook on the looseness of theoretical privacy bounds for DP-SGD, the de facto private training method in modern ML, as nearly-matching empirical lower bounds have been achieved under various threat models [NHSBTJCT23, AC24, CBP25]. In this work, we propose the empirical privacy lower bound of an algorithm as a concrete metric to optimize for, complementary to the theoretical upper bound. We give a lightweight defense framework that generically augments optimization methods in the ML pipeline to have significantly-improved empirical privacy on standard benchmarks. Moreover, we show that our framework comes at no theoretical privacy cost when augmenting DP-SGD, unlike previously-proposed defenses against membership inference attacks. We evaluate our defense against a broad range of audit constructions, models, and datasets to demonstrate its flexibility.

── more in #ai-safety 4 stories · sorted by recency
── more on @arxiv 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/revisiting-the-prova…] indexed:0 read:1min 2026-09-01 ·