Picture for Qingyu Huang

Qingyu Huang

Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage

Add code
Feb 07, 2025
Figure 1 for Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage
Figure 2 for Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage
Figure 3 for Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage
Figure 4 for Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage
Viaarxiv icon

Privacy Token: Surprised to Find Out What You Accidentally Revealed

Add code
Feb 06, 2025
Figure 1 for Privacy Token: Surprised to Find Out What You Accidentally Revealed
Figure 2 for Privacy Token: Surprised to Find Out What You Accidentally Revealed
Figure 3 for Privacy Token: Surprised to Find Out What You Accidentally Revealed
Figure 4 for Privacy Token: Surprised to Find Out What You Accidentally Revealed
Viaarxiv icon

Intermediate Outputs Are More Sensitive Than You Think

Add code
Dec 01, 2024
Figure 1 for Intermediate Outputs Are More Sensitive Than You Think
Figure 2 for Intermediate Outputs Are More Sensitive Than You Think
Figure 3 for Intermediate Outputs Are More Sensitive Than You Think
Figure 4 for Intermediate Outputs Are More Sensitive Than You Think
Viaarxiv icon

Enhancing DP-SGD through Non-monotonous Adaptive Scaling Gradient Weight

Add code
Nov 05, 2024
Viaarxiv icon

Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios

Add code
Jun 24, 2024
Figure 1 for Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios
Figure 2 for Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios
Figure 3 for Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios
Viaarxiv icon