Differentially private dropout technique preserves privacy in neural network training.
problem Preserving privacy in large datasets used for neural network training.
method Introduces a Bayesian dropout technique that adds intrinsic noise for regularization and differential privacy.
result Demonstrates that the iterative nature of neural network training can be handled with a relaxed differential privacy concept.
PGFL framework learns personalized models with differential privacy.
problem Privacy-preserving personalized learning for diverse data.
method Exploits model similarities and differential privacy (zero-concentrated).
result Algorithm converges to optimal solutions with linear time complexity.
Paper addresses DP in bandits, focusing on zCDP and providing private algorithms.
problem Privacy concerns in recommender systems using user-sensitive data.
method Formalizes and compares different DP adaptations to bandits, proposes private algorithms for various bandit settings.
result Private algorithms ensure negligible privacy costs compared to non-private regret.
Improved privacy-preserving methods for convex optimization with heavy-tailed data.
problem Privacy-preserving optimization of convex functions with heavy-tailed data.
method Developed algorithms for private mean estimation and convex optimization under concentrated differential privacy constraints.
result Achieved improved upper bounds on excess population risk for convex and strongly convex loss functions.
The paper develops methods to create confidence intervals for machine learning models while maintaining differential privacy.
problem Designing confidence intervals for differentially private machine learning models.
method Develops algorithms to create confidence intervals that satisfy differential privacy and concentrated differential privacy.
result The algorithms can be used with existing differentially private mechanisms that train models using objective perturbation and output perturbation.
Develops privacy-preserving multivariate median estimation methods.
problem Lack of rigorous privacy guarantees for robust multivariate location estimation.
method Novel finite-sample performance guarantees for differentially private multivariate depth-based medians.
result Sharp performance guarantees for multivariate depth-based medians under differential privacy.
New filters match advanced composition for adaptive privacy, with practical constants.
problem Limitations of existing adaptive composition methods.
method Constructed new filters and odometers that match advanced composition rates, including constants.
result Achieved fully adaptive privacy with practical filters and odometers.
Differentially private GAN model protects sensitive data privacy.
problem GANs can remember training samples, leaking sensitive data.
method Add noise to gradients during learning to achieve differential privacy.
result DPGAN generates high quality data points at a reasonable privacy level.
Paper proposes P-ADMM for ADMM in distributed medical machine learning with differential privacy.
problem Privacy leakage in ADMM for distributed machine learning with sensitive data.
method Integrates Gaussian noise with linearly decaying variance to provide dynamic zCDP.
result P-ADMM achieves the same convergence rate as non-private ADMM while ensuring differential privacy.
Paper develops a federated learning method to protect privacy without sacrificing model utility.
problem Privacy leakage in federated learning due to information exchange between edge devices and server.
method Combines local gradient perturbation, secure aggregation, and zCDP for privacy protection.
result Demonstrates superior trade-off between privacy and model utility through extensive experiments.
New method for estimating median and mean with high probability privacy.
problem Estimating median and mean with differential privacy.
method Propose, Test, Release (PTR) mechanism with concentration inequalities.
result First sub-Gaussian high probability bounds for differentially private median and mean estimation.
Private statistics estimation faces a bias, accuracy, and privacy trilemma.
problem Balancing privacy, accuracy, and bias in statistical estimation.
method Use differential privacy (DP) for private statistics, but clip samples to control sensitivity and add noise for privacy, introducing bias.
result No algorithm can simultaneously have low bias, low error, and low privacy loss for arbitrary distributions.
Privacy constraints affect learning Markov Random Fields differently.
problem Learning Markov Random Fields under differential privacy constraints.
method Algorithms for structure and parameter learning under pure, concentrated, and approximate differential privacy.
result Privacy constraints impose a strong separation between structure and parameter learning in high-dimensional data.
New algorithm protects privacy in IVaR regression while maintaining accuracy.
problem Privacy leakage in classical IVaR methods.
method Noisy two-stage gradient descent with differential privacy guarantees.
result Achieves statistical efficiency and privacy in IVaR regression.
Privacy-preserving dropout technique for neural networks.
problem Overfitting in deep neural networks and privacy of training data.
method Modified variational dropout with concentrated differential privacy.
result The technique provides differential privacy for neural networks.
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.
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 gradient descent for privacy with adaptive privacy budget.
problem Creating differentially private gradient descent algorithms.
method Adaptive allocation of privacy budget per iteration, with line-search capability.
result Outperforms prior algorithms for model fitting and is competitive for (ε,δ)-differential privacy. Iteratively reweighted least squares (IRLS) is a widely-used method in machine learning to estimate the parameters in the generalised linear models. In particular, IRLS for L1 minimisation under the linear model provides a closed-form solution in each step, which is a simple multiplication between the inverse of the we…
The iterative nature of the expectation maximization (EM) algorithm presents a challenge for privacy-preserving estimation, as each iteration increases the amount of noise needed. We propose a practical private EM algorithm that overcomes this challenge using two innovations: (1) a novel moment perturbation formulation…
New algorithms protect user-level privacy in learning tasks.
problem Protecting user-level privacy in learning tasks with stringent constraints.
method Proposes algorithms for learning tasks under user-level differential privacy constraints.
result Privacy cost decreases as users provide more samples or as the number of users increases.
PPGAN adds noise to GAN gradients to protect private data.
problem Protecting private data in GANs when training on sensitive datasets.
method Integrates differential privacy into GANs by adding noise to gradients and using Moments Accountant strategy.
result Demonstrates generation of high-quality synthetic data while maintaining privacy.
Improved regret bounds for DP-KLUCB and DP-IMED in Bernoulli bandits.
problem Minimizing regret in stochastic bandits under ε-global Differential Privacy.
method Developed DP versions of KLUCB and IMED, proving tighter lower bounds and matching upper bounds.
result DP-KLUCB and DP-IMED achieve asymptotically optimal regret under ε-global DP.
PATE scales to large-scale learning tasks with improved privacy and utility.
problem Privacy concerns in machine learning models trained on sensitive data.
method Private Aggregation of Teacher Ensembles (PATE) with new noisy aggregation mechanisms.
result Improved scalability and stronger privacy guarantees for PATE.
Optimal Gaussian noise mechanisms achieve nearly optimal error in unbiased mean estimation.
problem Efficiently estimating the mean of high-dimensional data while preserving privacy.
method Differential privacy mechanisms with Gaussian noise, focusing on optimal covariance.
result Gaussian noise mechanisms achieve nearly optimal error among all private unbiased mean estimation mechanisms.
Polynomial-time algorithm estimates mean with bounded covariance using differential privacy.
problem Estimating mean of a d-variate distribution with differential privacy constraints.
method Sum of Squares (SoS) exponential mechanism for polynomial-time differentially private estimation.
result First polynomial-time algorithm with O(d) samples for mean estimation under pure differential privacy. Novel PP-ADMM and IPP-ADMM algorithms improve differential privacy in distributed machine learning.
problem Privacy concerns in ADMM-based distributed machine learning.
method Proposes PP-ADMM and IPP-ADMM algorithms to provide differential privacy while improving model accuracy and convergence.
result The proposed algorithms achieve better model accuracy and convergence under the same privacy guarantee.
PACE-GGM uses Gaussian mechanism for private covariance estimation.
problem Private estimation of covariance matrices in high dimensions.
method Data-adaptive selection of entries, Gaussian mechanism, maximum-entropy reconstruction.
result Consistent improvements in estimation error compared to Gaussian mechanism and baselines.
Unified bounds for DP risks reduce noise and improve accuracy.
problem Difficult interpretation and calibration of DP mechanisms.
method Hypothesis-testing interpretation of DP (f-DP) and unified bounds. result Unified bounds are tighter and tunable for specific risks.
Private Generative Bootstrap protects privacy in statistical reporting.
problem Protecting privacy in statistical reporting of individual data.
method Bayesian likelihood-free framework with blocking strategy for differential privacy.
result Private Generative Bayesian Bootstrap (PGBB) provides competitive uncertainty quantification.
Optimizes privacy-preserving optimization for heavy-tailed data.
problem Privacy-preserving optimization with heavy-tailed gradients.
method Pure ε-differential privacy framework for Lipschitz extensions.
result Minimax optimal excess-risk rate for pure ε-DP heavy-tailed SCO.
Develops a computationally tractable differentially private mean estimator called the balloon mean.
problem Robust mean estimation in the presence of outliers and heavy-tailed distributions.
method Iterative clipping procedure over Mahalanobis balls.
result Balloon mean is robust to outliers and outperforms existing estimators in contaminated settings.
Noise addition prevents overfitting in adaptive data analysis.
problem Overfitting in repeated use of a data sample via adaptively chosen queries.
method Simple noise addition algorithms and differential privacy-based analysis.
result Noise-addition algorithms provide variance-dependent guarantees for unbounded queries.
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.
New privacy framework tailored to specific data distributions.
problem Protecting individual data points in decision-making processes.
method Introducing tangent differential privacy, a new form of differential privacy.
result Entropic regularization guarantees tangent differential privacy under general conditions.
Differential privacy is a statistical concept that can be explained through hypothesis testing.
problem Formalizing differential privacy as a statistical concept.
method Using David Blackwell's informativeness theorem, the paper shows differential privacy can be understood through hypothesis testing.
result The definition of f-differential privacy provides a unified framework for analyzing privacy bounds. Improved BAI under DP reduces gap to constant.
problem Fixed-confidence BAI under global DP for Bernoulli distributions.
method New lower bound, stopping rule, and Top Two sampling rule.
result Reduces gap to a small multiplicative constant.
New federated f-differential privacy for collaborative learning.
problem Privacy in federated learning.
method Introducing federated f-differential privacy and proposing a generic private federated learning framework. result Proves federated f-differential privacy provides privacy guarantee on each record of one client's data. Addresses theoretical and practical aspects of Gaussian differential privacy.
problem Theoretical and practical challenges in privacy-preserving data analysis.
method Discussion of f-differential privacy and Gaussian differential privacy.
result Gaussian differential privacy can enhance privacy in various applications.
Federated learning with Bayesian differential privacy offers improved privacy and accuracy.
problem Privacy in federated learning with similar data distributions.
method Bayesian differential privacy for federated learning, with improved privacy budgeting.
result Significant advantage over state-of-the-art privacy bounds, with lower noise and improved accuracy.
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.
Differential privacy improves AI security, fairness, and learning.
problem Privacy violations, security issues, and model fairness in AI.
method Application of differential privacy in various AI areas.
result Differential privacy enhances AI performance in multiple areas.
Three tools improve practical differential privacy.
problem Challenges in applying differential privacy to real-world data.
method Three tools: sanity checks, adaptive clipping, and large-batch training.
result Improves model performance and practicality of differential privacy.
New method reduces privacy impact on model accuracy for underrepresented groups.
problem Privacy mechanisms disproportionately affect underrepresented groups in machine learning models.
method Proposes DPSGD-F, a modified DPSGD that adjusts group contributions based on clipping bias.
result DPSGD-F removes disparate impact of differential privacy on model accuracy for protected groups.
This paper benchmarks privacy-preserving machine learning on medical images.
problem Ensuring privacy in medical image analysis while maintaining model accuracy.
method Comparing Local-DP and DP-SGD for differential privacy in medical imagery.
result Theoretical privacy guarantees do not fully align with real-world performance.
Novel compression method preserves privacy while reducing communication costs.
problem Reducing communication costs in differential privacy mechanisms.
method Poisson private representation (PPR) for compressing and simulating local randomizers.
result Achieves compression within a logarithmic gap from theoretical lower bound.
Differential privacy protects data privacy by adding noise to data.
problem Leakage of sensitive data through common methods like encryption and endpoint protection.
method Randomized response technique to add noise to data collection.
result Differential privacy ensures strong privacy with better utility.
Paper introduces differential privacy for sparse classification learning.
problem Privacy-preserving sparse classification learning.
method Differential privacy via ADMM with exponential noise addition.
result Proposes a privacy-preserving logistic regression algorithm.