Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD

• URL: http://arxiv.org/abs/2601.10237v1
• Date: Thu, 15 Jan 2026 09:50:36 GMT
• Title: Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD
• Authors: Murat Bilgehan Ertan, Marten van Dijk,
• Abstract summary: We analyze Differentially Private Gradient Descent (DP-SGD) in the $f$differential privacy framework.<n>We prove that enforcing a small separation imposes a strict lower bound on the noise multiplier $$, which directly limits the achievable utility.<n>Our experiments confirm that the noise levels implied by this bound leads to significant accuracy degradation at realistic training settings.
• Score: 7.787109481104569
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Differentially Private Stochastic Gradient Descent (DP-SGD) is the dominant paradigm for private training, but its fundamental limitations under worst-case adversarial privacy definitions remain poorly understood. We analyze DP-SGD in the $f$-differential privacy framework, which characterizes privacy via hypothesis-testing trade-off curves, and study shuffled sampling over a single epoch with $M$ gradient updates. We derive an explicit suboptimal upper bound on the achievable trade-off curve. This result induces a geometric lower bound on the separation $κ$ which is the maximum distance between the mechanism's trade-off curve and the ideal random-guessing line. Because a large separation implies significant adversarial advantage, meaningful privacy requires small $κ$. However, we prove that enforcing a small separation imposes a strict lower bound on the Gaussian noise multiplier $σ$, which directly limits the achievable utility. In particular, under the standard worst-case adversarial model, shuffled DP-SGD must satisfy $σ\ge \frac{1}{\sqrt{2\ln M}}$ $\quad\text{or}\quad$ $κ\ge\ \frac{1}{\sqrt{8}}\!\left(1-\frac{1}{\sqrt{4π\ln M}}\right)$, and thus cannot simultaneously achieve strong privacy and high utility. Although this bound vanishes asymptotically as $M \to \infty$, the convergence is extremely slow: even for practically relevant numbers of updates the required noise magnitude remains substantial. We further show that the same limitation extends to Poisson subsampling up to constant factors. Our experiments confirm that the noise levels implied by this bound leads to significant accuracy degradation at realistic training settings, thus showing a critical bottleneck in DP-SGD under standard worst-case adversarial assumptions.

Related papers

• Can SGD Handle Heavy-Tailed Noise? [6.111519084375339]
  Gradient Descent (SGD) is a machine learning project of large-scale optimization, yet its theoretical behavior under heavy-tailed noise is poorly understood.<n>We rigorously investigate whether SGD, can provably succeed under such adverse conditions.
  arXiv  Detail & Related papers  (2025-08-06T20:09:41Z)
• Better Rates for Private Linear Regression in the Proportional Regime via Aggressive Clipping [19.186034457189162]
  A common approach is to set the clipping constant much larger than the expected norm of the per-sample gradients.<n>While simplifying the analysis, this is however in sharp contrast with what empirical evidence suggests to optimize performance.<n>Our work bridges this gap between theory and practice by crucially operating in a regime where clipping happens frequently.
  arXiv  Detail & Related papers  (2025-05-22T07:34:27Z)
• Smoothed Normalization for Efficient Distributed Private Optimization [54.197255548244705]
  Federated learning enables machine learning models with privacy of participants.<n>There is no differentially private distributed method for training, non-feedback problems.<n>We introduce a new distributed algorithm $alpha$-$sf NormEC$ with provable convergence guarantees.
  arXiv  Detail & Related papers  (2025-02-19T07:10:32Z)
• Beyond Covariance Matrix: The Statistical Complexity of Private Linear Regression [66.93988594607842]
  Under privacy constraints, the complexity of private linear regression is emphnot captured by the usual covariance matrix.<n>We introduce an Information-Weighted Regression method that attains the optimal rates.<n> Notably, our results demonstrate that joint privacy comes at almost no additional cost.
  arXiv  Detail & Related papers  (2025-02-18T18:35:24Z)
• Differentially Private Best-Arm Identification [14.916947598339988]
  Best Arm Identification (BAI) problems are progressively used for data-sensitive applications. Motivated by the data privacy concerns invoked by these applications, we study the problem of BAI with fixed confidence in both the local and central models.
  arXiv  Detail & Related papers  (2024-06-10T16:02:48Z)
• A Generalized Shuffle Framework for Privacy Amplification: Strengthening Privacy Guarantees and Enhancing Utility [4.7712438974100255]
  We show how to shuffle $(epsilon_i,delta_i)$-PLDP setting with personalized privacy parameters. We prove that shuffled $(epsilon_i,delta_i)$-PLDP process approximately preserves $mu$-Gaussian Differential Privacy with mu = sqrtfrac2sum_i=1n frac1-delta_i1+eepsilon_i-max_ifrac1-delta_i1+e
  arXiv  Detail & Related papers  (2023-12-22T02:31:46Z)
• Private Online Prediction from Experts: Separations and Faster Rates [74.52487417350221]
  Online prediction from experts is a fundamental problem in machine learning and several works have studied this problem under privacy constraints. We propose and analyze new algorithms for this problem that improve over the regret bounds of the best existing algorithms for non-adaptive adversaries.
  arXiv  Detail & Related papers  (2022-10-24T18:40:19Z)
• Normalized/Clipped SGD with Perturbation for Differentially Private Non-Convex Optimization [94.06564567766475]
  DP-SGD and DP-NSGD mitigate the risk of large models memorizing sensitive training data. We show that these two algorithms achieve similar best accuracy while DP-NSGD is comparatively easier to tune than DP-SGD.
  arXiv  Detail & Related papers  (2022-06-27T03:45:02Z)
• Non-Euclidean Differentially Private Stochastic Convex Optimization [15.302167005107135]
  We show that noisy gradient descent (SGD) algorithms attain the optimal excess risk in low-dimensional regimes. Our work draws upon concepts from the geometry of normed spaces, such as the notions of regularity, uniform convexity, and uniform smoothness.
  arXiv  Detail & Related papers  (2021-03-01T19:48:44Z)
• Learning with User-Level Privacy [61.62978104304273]
  We analyze algorithms to solve a range of learning tasks under user-level differential privacy constraints. Rather than guaranteeing only the privacy of individual samples, user-level DP protects a user's entire contribution. We derive an algorithm that privately answers a sequence of $K$ adaptively chosen queries with privacy cost proportional to $tau$, and apply it to solve the learning tasks we consider.
  arXiv  Detail & Related papers  (2021-02-23T18:25:13Z)
• Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training [12.386462516398469]
  Finding efficient, easily implementable differentially private (DP) algorithms that offer strong excess risk bounds is an important problem in modern machine learning. We provide the tightest known $(epsilon, 0)$-DP population loss bounds and fastest runtimes under the presence of smoothness and strong convexity. We apply our theory to two learning frameworks: tilted ERM and adversarial learning frameworks.
  arXiv  Detail & Related papers  (2021-02-09T08:47:06Z)
• Private Stochastic Non-Convex Optimization: Adaptive Algorithms and Tighter Generalization Bounds [72.63031036770425]
  We propose differentially private (DP) algorithms for bound non-dimensional optimization. We demonstrate two popular deep learning methods on the empirical advantages over standard gradient methods.
  arXiv  Detail & Related papers  (2020-06-24T06:01:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.