Jalaj Upadhyay

activity

20232025

most citedOptimality of Matrix Mechanism on

ℓ_{p}^{p}

-metric

1 citations · 2 across the 6 of their papers we have counts for

collaborators

Showing cs.CRShow all

3 papers · 1 filter

cs.CR2025★ 1 cited

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD

Nikita P. Kalinin, Ryan McKenna, Jalaj Upadhyay +1

Matrix factorization mechanisms for differentially private training have emerged as a promising approach to improve model utility under privacy constraints. In practical settings,…

cs.CR2024

Almost linear time differentially private release of synthetic graphs

Jingcheng Liu, Jalaj Upadhyay, Zongrui Zou

In this paper, we give an almost linear time and space algorithms to sample from an exponential mechanism with an $ℓ_{1}$ -score function defined over an exponentially large non-co…

cs.CR2024★ 1 cited

Optimality of Matrix Mechanism on $ℓ_{p}^{p}$ -metric

Jingcheng Liu, Jalaj Upadhyay, Zongrui Zou

In this paper, we introduce the $ℓ_{p}^{p}$ -error metric (for $p \geq 2$ ) when answering linear queries under the constraint of differential privacy. We characterize such an error u…

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD

Almost linear time differentially private release of synthetic graphs

Optimality of Matrix Mechanism on ℓpp​-metric

Optimality of Matrix Mechanism on $ℓ_{p}^{p}$ -metric