An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees

• URL: http://arxiv.org/abs/2605.20521v1
• Date: Tue, 19 May 2026 21:43:22 GMT
• Title: An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees
• Abstract summary: Fine-tuning a pretrained machine learning model to a small, sensitive dataset risks memorizing individual new data points.<n>We develop a randomized algorithm based on the exponential mechanism for fine-tuning while ensuring differential privacy.<n> Numerical experiments on the MNIST benchmark and the MIMIC clinical dataset demonstrate competitive performance against existing differentially private fine-tuning techniques.
• Score: 6.856118376541429
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Fine-tuning adapts a pretrained machine learning model to a small, sensitive dataset, but this process risks memorizing individual new data points, making the model vulnerable to adversaries who seek to extract sensitive information. In this work, we develop a randomized algorithm based on the exponential mechanism for fine-tuning while ensuring differential privacy. Our key idea is to construct a simple utility function that combines a local quadratic approximation of the pretrained model with information from the new dataset. The resulting exponential mechanism admits exact sampling from a multivariate normal distribution in closed form. We establish theoretical privacy guarantees, sensitivity bounds, and accuracy estimations for our method. We further introduce a random-projection strategy that makes the approach scalable to high-dimensional models. Numerical experiments on the MNIST benchmark and the MIMIC clinical dataset demonstrate competitive performance against existing differentially private fine-tuning techniques.

Related papers

• Phantoms and Disclosures: a Causal Framework for Auditing Synthetic Data [56.86147283213652]
  We present a customizable empirical auditing framework designed to detect and explain data disclosures.<n>Our framework introduces a mechanism to distinguish between "true disclosures"-where the system directly reproduces a user's information-and "phantom disclosures"-where the system incidentally generates a user's data.
  arXiv  Detail & Related papers  (2026-06-15T16:54:02Z)
• Joint Model and Data Sparsification via the Marginal Likelihood [53.29070892356214]
  We propose jointly learning individual feature and sample relevancies, enabling simultaneous model and data sparsification.<n>This symmetric pruning of model and data offers a natural extension that preserves conjugacy.<n> Empirical results across diverse regression tasks affirm that a joint ARD approach consistently yields both sparse and robust prediction models.
  arXiv  Detail & Related papers  (2026-05-28T13:26:53Z)
• High-Dimensional Differentially Private Quantile Regression: Distributed Estimation and Statistical Inference [0.26784722398800515]
  We propose a differentially private quantile regression method for high-dimensional data in a distributed setting.<n>We develop a differentially private estimation algorithm with iterative updates, ensuring near-optimal statistical accuracy and formal privacy guarantees.
  arXiv  Detail & Related papers  (2025-08-07T09:47:44Z)
• Private Training & Data Generation by Clustering Embeddings [74.00687214400021]
  Differential privacy (DP) provides a robust framework for protecting individual data.<n>We introduce a novel principled method for DP synthetic image embedding generation.<n> Empirically, a simple two-layer neural network trained on synthetically generated embeddings achieves state-of-the-art (SOTA) classification accuracy.
  arXiv  Detail & Related papers  (2025-06-20T00:17:14Z)
• Differentially Private Random Feature Model [47.35176457481132]
  We produce a differentially private random feature model for privacy-preserving kernel machines.<n>We show that our method preserves privacy and derive a generalization error bound for the method.
  arXiv  Detail & Related papers  (2024-12-06T05:31:08Z)
• Bayesian Estimation and Tuning-Free Rank Detection for Probability Mass Function Tensors [17.640500920466984]
  This paper presents a novel framework for estimating the joint PMF and automatically inferring its rank from observed data. We derive a deterministic solution based on variational inference (VI) to approximate the posterior distributions of various model parameters. Additionally, we develop a scalable version of the VI-based approach by leveraging variational inference (SVI) Experiments involving both synthetic data and real movie recommendation data illustrate the advantages of our VI and SVI-based methods in terms of estimation accuracy, automatic rank detection, and computational efficiency.
  arXiv  Detail & Related papers  (2024-10-08T20:07:49Z)
• Distributed Personalized Empirical Risk Minimization [19.087524494290676]
  This paper advocates a new paradigm Personalized Empirical Risk Minimization (PERM) to facilitate learning from heterogeneous data sources. We propose a distributed algorithm that replaces the standard model averaging with model shuffling to simultaneously optimize PERM objectives for all devices.
  arXiv  Detail & Related papers  (2023-10-26T20:07:33Z)
• On the Computational Complexity of Private High-dimensional Model Selection [18.964255744068122]
  We consider the problem of model selection in a high-dimensional sparse linear regression model under privacy constraints. We propose an efficient Metropolis-Hastings algorithm and under certain regularity conditions, we establish that it enjoys mixing time to its stationary distribution.
  arXiv  Detail & Related papers  (2023-10-11T19:53:15Z)
• Self-Supervised Dataset Distillation for Transfer Learning [77.4714995131992]
  We propose a novel problem of distilling an unlabeled dataset into a set of small synthetic samples for efficient self-supervised learning (SSL) We first prove that a gradient of synthetic samples with respect to a SSL objective in naive bilevel optimization is textitbiased due to randomness originating from data augmentations or masking. We empirically validate the effectiveness of our method on various applications involving transfer learning.
  arXiv  Detail & Related papers  (2023-10-10T10:48:52Z)
• HyperImpute: Generalized Iterative Imputation with Automatic Model Selection [77.86861638371926]
  We propose a generalized iterative imputation framework for adaptively and automatically configuring column-wise models. We provide a concrete implementation with out-of-the-box learners, simulators, and interfaces.
  arXiv  Detail & Related papers  (2022-06-15T19:10:35Z)

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.