Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Spectrum-Aware Bounds on Invertibility for Privacy-Enhancing Instance Encoding
Seokjin Hwang, Yuting Li, Yuting +2
Instance encoding is a popular empirical technique for privacy enhancement when sharing data to an untrusted server. It transforms sensitive data through an encoding process before…
cs.LG2026
Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
Xin Gu, Yingtai Xiao, Guanlin He +3
Differentially private stochastic gradient descent (DP-SGD) offers the promise of training deep learning models while mitigating many privacy risks. However, there is currently a l…
cs.LG2024
Information Flow Control in Machine Learning through Modular Model Architecture
Trishita Tiwari, Suchin Gururangan, Chuan Guo +7
In today's machine learning (ML) models, any part of the training data can affect the model output. This lack of control for information flow from training data to model output is…