Picture for Andrew Lowy

Andrew Lowy

Efficient Differentially Private Fine-Tuning of Diffusion Models

Add code
Jun 07, 2024
Viaarxiv icon

How to Make the Gradients Small Privately: Improved Rates for Differentially Private Non-Convex Optimization

Feb 17, 2024
Viaarxiv icon

Why Does Differential Privacy with Large Epsilon Defend Against Practical Membership Inference Attacks?

Feb 14, 2024
Viaarxiv icon

Optimal Differentially Private Learning with Public Data

Add code
Jun 26, 2023
Figure 1 for Optimal Differentially Private Learning with Public Data
Figure 2 for Optimal Differentially Private Learning with Public Data
Figure 3 for Optimal Differentially Private Learning with Public Data
Figure 4 for Optimal Differentially Private Learning with Public Data
Viaarxiv icon

Stochastic Differentially Private and Fair Learning

Add code
Oct 17, 2022
Figure 1 for Stochastic Differentially Private and Fair Learning
Figure 2 for Stochastic Differentially Private and Fair Learning
Figure 3 for Stochastic Differentially Private and Fair Learning
Figure 4 for Stochastic Differentially Private and Fair Learning
Viaarxiv icon

Private Stochastic Optimization in the Presence of Outliers: Optimal Rates for (Non-Smooth) Convex Losses and Extension to Non-Convex Losses

Sep 15, 2022
Figure 1 for Private Stochastic Optimization in the Presence of Outliers: Optimal Rates for (Non-Smooth) Convex Losses and Extension to Non-Convex Losses
Viaarxiv icon

Private Non-Convex Federated Learning Without a Trusted Server

Add code
Mar 13, 2022
Figure 1 for Private Non-Convex Federated Learning Without a Trusted Server
Figure 2 for Private Non-Convex Federated Learning Without a Trusted Server
Figure 3 for Private Non-Convex Federated Learning Without a Trusted Server
Figure 4 for Private Non-Convex Federated Learning Without a Trusted Server
Viaarxiv icon

Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds

Add code
Jun 17, 2021
Figure 1 for Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds
Figure 2 for Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds
Figure 3 for Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds
Figure 4 for Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds
Viaarxiv icon

FERMI: Fair Empirical Risk Minimization via Exponential Rényi Mutual Information

Add code
Feb 24, 2021
Figure 1 for FERMI: Fair Empirical Risk Minimization via Exponential Rényi Mutual Information
Figure 2 for FERMI: Fair Empirical Risk Minimization via Exponential Rényi Mutual Information
Figure 3 for FERMI: Fair Empirical Risk Minimization via Exponential Rényi Mutual Information
Figure 4 for FERMI: Fair Empirical Risk Minimization via Exponential Rényi Mutual Information
Viaarxiv icon

Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training

Feb 09, 2021
Figure 1 for Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training
Figure 2 for Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training
Figure 3 for Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training
Figure 4 for Output Perturbation for Differentially Private Convex Optimization with Improved Population Loss Bounds, Runtimes and Applications to Private Adversarial Training
Viaarxiv icon