New divergences help audit DP in high dimensions.
problem Challenges in auditing DP in high-dimensional data.
method Propose kernel Rényi divergence and its regularized version for auditing.
result Regularized kernel Rényi divergence can be estimated from samples in high dimensions.
Improves privacy amplification by shuffling for differential privacy.
problem Enhancing privacy guarantees in systems with anonymous data contributions.
method Theoretical and numerical analysis of Rényi differential privacy parameters and privacy amplification by shuffling.
result First asymptotically optimal analysis of Rényi differential privacy parameters for shuffled outputs.
Improved privacy bounds enhance deep learning training efficiency.
problem Enhancing privacy guarantees in deep learning models.
method Deriving optimal DP parameters using f-divergences. result Significantly reduces the number of iterations needed for training deep learning models.
Paper develops novel privacy mechanism for Riemannian manifold data using geometric analysis and heat diffusion.
problem Privacy-preserving estimation of generalized Frechet mean on Riemannian manifolds.
method Characterizes Renyi divergence via Harnack inequalities, introduces mechanisms based on heat diffusion and Langevin process.
result Proposes mechanisms for nonnegative and general Riemannian manifolds with detailed utility analyses.
A new privacy accountant for Gaussian differential privacy measures individual privacy losses.
problem Bounding differential privacy loss for each participant in data analysis.
method Developed a privacy accountant for adaptive compositions of randomised mechanisms using Gaussian differential privacy.
result Provided optimal bounds for the Gaussian mechanism and constructed an approximative individual privacy accountant.
Enhances privacy in machine learning through Rényi Pufferfish mechanisms.
problem Designing general and efficient Pufferfish mechanisms that maintain privacy and utility.
method Introduces a Rényi divergence-based variant of Pufferfish, generalizes the Wasserstein mechanism, and proves privacy amplification results.
result Extends the applicability of Pufferfish framework and provides stronger privacy guarantees.
Paper simplifies DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
problem Tension between efficiency and flexibility in DP composition theorems.
method Rényi Differential Privacy (RDP) for adaptive privacy budgets, proving simpler composition theorem with smaller constants.
result Practical DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
The Sampled Gaussian Mechanism (SGM)---a composition of subsampling and the additive Gaussian noise---has been successfully used in a number of machine learning applications. The mechanism's unexpected power is derived from privacy amplification by sampling where the privacy cost of a single evaluation diminishes quadr…
Differential privacy is a useful tool to build machine learning models which do not release too much information about the training data. We study the Rényi differential privacy of stochastic gradient descent when each training example is sampled without replacement (also known as cyclic SGD). Cyclic SGD is typically f…
We study the problem of subsampling in differential privacy (DP), a question that is the centerpiece behind many successful differentially private machine learning algorithms. Specifically, we provide a tight upper bound on the Rényi Differential Privacy (RDP) (Mironov, 2017) parameters for algorithms that: (1) subsamp…
Differential privacy is a de facto standard in data privacy, with applications in the public and private sectors. A way to explain differential privacy, which is particularly appealing to statistician and social scientists is by means of its statistical hypothesis testing interpretation. Informally, one cannot effectiv…
A new method for tighter privacy loss accounting in adaptive analyses.
problem Ensuring individual privacy in adaptive analyses while staying within a privacy budget.
method A personalized privacy loss estimate and a Rényi differential privacy filter.
result Personalized privacy loss accounting can be practical and tighter than existing methods.
New method makes CNN interpretations robust to adversarial attacks.
problem Adversarial attacks on CNN interpretation maps.
method Renyi Differential Privacy (RDP) for robust interpretation.
result Certifiable top-k robustness and improved experimental robustness. Study on privacy leakage in noisy gradient descent algorithms.
problem Information leakage of iterative randomized learning algorithms about training data.
method Analyzes the dynamics of Rényi differential privacy loss in noisy gradient descent algorithms.
result Privacy loss converges exponentially fast for smooth and strongly convex loss functions.
Improved privacy analysis for stochastic gradient descent.
problem Analyzing privacy leakage in noisy stochastic gradient descent.
method Modeling Rényi divergence dynamics with Langevin diffusions, proving exponential privacy loss convergence for smooth and strongly convex objectives.
result Privacy loss converges exponentially fast for smooth and strongly convex objectives under constant step size.
This short note highlights some links between two lines of research within the emerging topic of trustworthy machine learning: differential privacy and robustness to adversarial examples. By abstracting the definitions of both notions, we show that they build upon the same theoretical ground and hence results obtained …
New method for privacy amplification without sampling for matrix factorization.
problem Privacy amplification for differentially private model training with matrix factorization.
method Sampling-free bounds based on Rényi divergence and conditional composition.
result Stronger privacy guarantees for small ε, applicable to various matrices.
New RDP guarantees for heavy-tailed SDEs and SGD.
problem Characterizing differential privacy for heavy-tailed noise in learning algorithms.
method Rényi flow computations and fractional Poincaré inequalities.
result First RDP guarantees for heavy-tailed SDEs with weaker dependence on dimension.
A new Gaussian mechanism for differential privacy in the shuffle model is introduced.
problem Improving differential privacy in distributed learning environments.
method Characterization and upper-bounding of Rényi differential privacy (RDP) for the shuffle Gaussian mechanism.
result The shuffle Gaussian mechanism provides improved privacy guarantees compared to existing methods.
New method optimizes privacy and compute trade-offs for deep learning.
problem Privacy and compute trade-offs in deep learning training.
method Decoupling privacy analysis and experimental behavior, using TAN and scaling laws for DP-SGD.
result Stronger privacy guarantees with significant reduction in computational budget.
Three DP variants linked, improving SGD privacy bounds.
problem Relating different DP variants for tighter privacy bounds.
method Developed machinery to relate approximate DP to RDP and hypothesis test DP.
result Improved privacy guarantees for noisy SGD.
The Dirichlet mechanism protects privacy while minimizing KL divergence.
problem Minimizing KL divergence while protecting sensitive data privacy.
method Using the exponential mechanism with the KL divergence loss function, resulting in the Dirichlet mechanism.
result Proved a probability tail bound on KL divergence and derived a lower bound for sample complexity.
Proposes a privacy-preserving recommendation system using matrix factorization and differential privacy.
problem Privacy leakage in recommendation systems when anonymizing user data is not sufficient.
method Uses matrix factorization and differential privacy via the Gaussian mechanism.
result Demonstrates excellent utility for privacy-preserving recommendation systems.
Locally private methods detect changes in time series data.
problem Detecting distributional changes in time series data under local differential privacy.
method Proposed locally differentially private algorithms based on randomized response and binary mechanisms.
result Theoretical performance bounds and empirical validation of detection accuracy.
In this paper we consider the problem of minimizing composite objective functions consisting of a convex differentiable loss function plus a non-smooth regularization term, such as L1 norm or nuclear norm, under Rényi differential privacy (RDP). To solve the problem, we propose two stochastic alternating direction m…
RDP-GAN improves GAN privacy by adding random noises to loss function.
problem Protecting sensitive information in GANs while maintaining quality of generated samples.
method Integrates Rényi-differential privacy into GAN training process by adding random noises to loss function.
result Achieves better privacy protection with high-quality samples compared to existing methods.
Differential privacy of Gaussian process posterior sampling
problem Privacy of posterior sample paths from Gaussian process
method Intrinsic randomness yields DP guarantees
result Intrinsic randomness yields DP guarantees
DP-CGAN generates private synthetic data and labels.
problem Preserving privacy in synthetic data generation.
method Differentially private conditional GAN (DP-CGAN) with clipping and perturbation.
result DP-CGAN generates visually and empirically promising results on MNIST with low privacy cost.
Constructs models to meet any DP requirement without retraining.
problem Evolving privacy requirements at inference time.
method Two post-processing techniques: random selection and linear combination.
result Final private models satisfying any target privacy parameter.
New method shows hidden state can significantly improve differential privacy in SGD.
problem Differential privacy in SGD with hidden state.
method Proves converging privacy bounds for hidden state SGD, using privacy amplification techniques.
result Privacy bound converges exponentially fast and is smaller than composition bounds.
Paper improves differential privacy analysis for machine learning.
problem Quantifying privacy leakage in noisy gradient descent.
method Shifted interpolation in f-differential privacy. result First exact privacy analysis for strongly convex optimization.
Paper improves privacy for language models against reconstruction attacks.
problem Reconstruction attacks can regenerate training data from language models.
method Uses Rényi differential privacy with optimized privacy budgets.
result Better privacy guarantees for extraction of rare secrets.
This paper improves privacy accounting in decentralized FL using f-Differential Privacy.
problem Challenges in accurately quantifying privacy budget in decentralized FL.
method Develops two new f-DP-based accounting methods for decentralized FL.
result Yields tighter (ε,δ) bounds and improved utility compared to existing methods.
Providing meaningful privacy to users of location based services is particularly challenging when multiple locations are revealed in a short period of time. This is primarily due to the tremendous degree of dependence that can be anticipated between points. We propose a Rényi differentially private framework for boundi…
Paper extends FFT-based differential privacy method to heterogeneous compositions.
problem Computing accurate differential privacy guarantees for mixed mechanisms.
method Uses Fast Fourier Transform (FFT) for error analysis and parameter selection.
result Provides tighter bounds for heterogeneous compositions compared to homogeneous cases.
The paper provides privacy guarantees for MCMC algorithms using Langevin dynamics.
problem Ensuring differential privacy in MCMC algorithms.
method Novel methodology combining Girsanov's theorem and perturbation trick.
result Established (Rényi) DP guarantees for Langevin algorithms.
Recent developments in differentially private (DP) machine learning and DP Bayesian learning have enabled learning under strong privacy guarantees for the training data subjects. In this paper, we further extend the applicability of DP Bayesian learning by presenting the first general DP Markov chain Monte Carlo (MCMC)…
PBM mechanism improves privacy and accuracy in federated learning.
problem Secure and private federated learning with limited privacy budget.
method Poisson Binomial mechanism for discrete differential privacy.
result Achieves same privacy-accuracy trade-offs as Gaussian mechanism.
A new federated learning framework with sparsification and adaptive optimization for privacy and efficiency.
problem Lack of sufficient privacy protection in federated learning.
method Integrates random sparsification with gradient perturbation and acceleration techniques to enhance privacy and efficiency.
result Outperforms previous differentially-private federated learning approaches in privacy and efficiency.
Improved shuffling technique amplifies privacy guarantees for anonymous data contributions.
problem Enhancing privacy in systems where data is contributed anonymously.
method Developed a new approach to random shuffling that amplifies differential privacy guarantees.
result Achieved asymptotically optimal privacy amplification with nearly optimal dependence in ε.
Unified framework for subsampling mechanisms with tighter privacy guarantees.
problem Improving privacy in machine learning models through subsampling.
method Conditional optimal transport for deriving mechanism-specific subsampling guarantees.
result Tighter privacy bounds for subsampled mechanisms compared to traditional methods.
A new discrete privacy mechanism for federated learning.
problem Differentially private federated learning with communication constraints.
method Skellam mechanism based on Poisson distributions.
result Skellam mechanism provides similar privacy-accuracy trade-offs as Gaussian mechanism.
New methods speed up training of differentially private deep learning models.
problem Training differentially private deep learning models is slower than non-private models.
method Derive and implement new per-example gradient clipping methods compatible with auto-differentiation.
result Significant training speed-ups (54x - 94x) for various models and architectures.
DP-SPRT improves privacy in sequential tests with near-optimal error rates.
problem Privacy constraints in sequential probability ratio tests.
method A wrapper for SPRT that uses a private mechanism to determine when to stop based on predefined intervals.
result DP-SPRT achieves near-optimal error rates and privacy guarantees.
Paper introduces a privacy-preserving line search method for optimization.
problem Optimization performance depends on step size tuning, which is difficult and privacy-sensitive.
method Introduces a stochastic adaptive line search algorithm that satisfies differential privacy.
result The algorithm efficiently uses privacy budget and outperforms existing private optimizers.
This work develops sampling methods for differential privacy using SHK geometry.
problem Approximating sampling for the exponential mechanism in differential privacy.
method Develops perturbation theory for SHK gradient flows and applies to differential privacy.
result Derives time-dependent Pure-DP guarantees and Approximate-DP certificates.
Paper develops privacy-preserving federated learning for nonsmooth objectives.
problem Solving nonsmooth objective functions in a privacy-preserving manner.
method Zero-concentrated differential privacy (zCDP) with Gaussian noise, distributed ADMM, and approximation of augmented Lagrangian.
result The algorithm achieves a competitive privacy-accuracy trade-off and converges to the exact solution.
Improved log-concave sampling to O(d1/2) with warm starts.
problem Sampling from strongly log-concave distributions efficiently.
method Warm starts and discretized underdamped Langevin diffusion.
result Achieved O(d1/2) complexity for high-accuracy sampling.