Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

3567111,0671,422 · Jun 202019922001200920172026
48 results for private models

Private learning of Gaussian Mixture Models without boundedness assumptions.

problem Private estimation of parameters of Gaussian Mixture Models with unbounded components.
method Reduction to non-private problem, blackbox privatization, Moitra and Valiant's algorithm.
result First sample complexity upper bound and polynomial time algorithm for privately learning GMMs.

Private learning can perform well in high dimensions, contrary to known results.

problem When does differentially private learning not suffer in high dimensions?
method Introduced a condition called restricted Lipschitz continuity to derive improved bounds for excess empirical and population risks.
result Gradients in private fine-tuning of large models are mostly controlled by a few principal components, similar to conditions for convex settings.

Algorithm selects public datasets for private machine learning.

problem Choosing the most suitable public dataset for private machine learning.
method Measures gradient subspace distance between public and private datasets.
result Excess risk scales with the subspace distance between gradients.

Public pretraining improves private model training even in extreme distribution shift scenarios.

problem Improving private model training accuracy in settings with large distribution shift.
method Empirical evaluation and theoretical explanation of public representations improving private training accuracy.
result Public representations can improve private training accuracy by up to 67% over private training from scratch in settings with large distribution shift.

Optimizes sparse fine-tuning for privacy in neural networks.

problem Performance gap between DP-SGD and non-private fine-tuning.
method Optimization-based approach using private gradient information for selecting trainable weights.
result Our selection method leads to better prediction accuracy compared to existing approaches.

Improved image generation with private data using perceptual features.

problem Difficulty in training generative models with differential privacy.
method Use pre-trained perceptual features to learn private data distribution.
result Generative models can generate high-quality images with low privacy budget (ϵ2\epsilon \approx 2).

Efficient private algorithms for estimating block models and mixture models.

problem Estimating block models and mixture models in high-dimensional settings.
method General tools for designing efficient private estimation algorithms.
result First efficient private algorithms for weak and exact recovery of stochastic block models.

Improved differentially private deep learning with group-wise clipping techniques.

problem Efficiency and privacy trade-offs in deep learning models.
method Group-wise clipping techniques (per-layer and per-device) to reduce compute time and memory overhead.
result Private learning with group-wise clipping achieves similar or better performance than non-private learning with less wall time.

We revisit the problem of differentially private release of classification queries. In this problem, the goal is to design an algorithm that can accurately answer a sequence of classification queries based on a private training set while ensuring differential privacy. We formally study this problem in the agnostic PAC …

2019-07-31abs ↗pdf ↗

DPZero fine-tunes large models privately without backpropagation.

problem Memory and privacy challenges in fine-tuning large language models.
method DPZero uses zeroth-order methods for private fine-tuning, avoiding backpropagation.
result DPZero achieves private fine-tuning of RoBERTa and OPT on various tasks.

Private method measures nonlinear correlations between data hosted across two entities.

problem Measuring nonlinear correlations between sensitive data hosted across multiple parties while preserving privacy.
method Differentially private estimator of distance correlation.
result First private estimator of nonlinear correlations in a multi-party setup.

New private learning algorithms improve utility in tasks with public features.

problem Private learning with public features in recommendation and ad prediction.
method Developed algorithms that protect only certain sufficient statistics, improving utility for linear regression and private recommendation benchmarks.
result Achieved state-of-the-art performance on private recommendation benchmarks.

Differentially private algorithms protect model explanations from leaking training data.

problem Model explanations can leak training data, compromising privacy.
method Adaptive differentially private gradient descent algorithm to produce accurate, private explanations.
result Privacy amplification and reduction of overall privacy loss on explanation data.

Differentially private conformal prediction improves statistical efficiency.

problem Quantifying uncertainty in private data analysis.
method Introducing differential conformal prediction and developing Differentially Private Conformal Prediction (DPCP).
result DPCP produces tighter prediction sets than existing private split conformal approaches.

We present new differentially private algorithms for learning a large-margin halfspace. In contrast to previous algorithms, which are based on either differentially private simulations of the statistical query model or on private convex optimization, the sample complexity of our algorithms depends only on the margin of…

2019-02-24abs ↗pdf ↗

New privacy-preserving learning model for mixtures of private and public data.

problem Learning from datasets with both private and public data, where privacy concerns differ.
method Designing a differential privacy-preserving learning algorithm for a mixture of private and public sub-populations.
result Linear classifiers can be learned with sample complexity comparable to non-private PAC-learning, even when privacy status correlates with labels.

DPlis improves privacy in deep learning models by smoothing loss functions.

problem Privacy leakage in deep learning models trained on private data and low model performance.
method DPlis constructs a smooth loss function to favor noise-resilient models.
result DPlis effectively boosts model quality and training stability under privacy constraints.

New algorithms for community detection in graphs with privacy constraints.

problem Community recovery in stochastic block models with node-wise privacy.
method Spectral clustering with privacy mechanisms, including privatized PCA, convex optimization, and matrix estimation.
result Developed algorithms that are computable in polynomial-time and achieve consistent community estimation under node differential privacy.

New algorithm learns regression models privately under growth condition.

problem Private learning of nonparametric regression models.
method Novel filtering procedure to output stable hypotheses for nonparametric function classes.
result Established first nonparametric private learnability guarantee for diverging fat shattering dimensions.

Discond-VAE separates continuous and discrete factors in data.

problem Separating shared and class-specific variations in real-world data.
method Introduces private and public latent variables to represent continuous and discrete factors, respectively.
result Discond-VAE successfully disentangles class-dependent continuous factors from discrete factors.

New algorithms for private generalized linear contextual bandits.

problem Private estimation and optimization for generalized linear models under differential privacy.
method Developed algorithms for stochastic and adversarial contexts under shuffle and joint differential privacy.
result Achieved private regret bounds for generalized linear models, differing from non-private rates by factors of d/ε\sqrt{d/\varepsilon} and d/ε\sqrt{d/\varepsilon} respectively.

Locally private reinforcement learning protects individual environments from reverse engineering.

problem Protecting private information in distributed reinforcement learning environments.
method Locally differentially private algorithms that protect local agents' models from adversarial reverse engineering.
result Demonstrated that the proposed algorithm performs well under local differential privacy (LDP).

Near-optimal private tests for simple and MLR hypotheses developed under Gaussian differential privacy.

problem Developing private tests for simple and MLR hypotheses under Gaussian differential privacy.
method A private mean estimator with data-driven clamping bounds, constructing private test statistics.
result Private tests achieve the same asymptotic relative efficiency as non-private most powerful tests.

Study private submodular maximization in streaming data.

problem Private maximization of submodular functions in streaming data.
method Established differentially private baselines and derived better trade-offs for decomposable submodular functions.
result Improved trade-offs between privacy and utility for decomposable submodular functions.