activity
20162021
most citedFast and Memory Efficient Differentially Private-SGD via JL Projections

8 citations · 8 across the 4 of their papers we have counts for

collaborators

7 papers

cs.LG2021

Differentially Private n-gram Extraction

Kunho Kim, Sivakanth Gopi, Janardhan Kulkarni +1

We revisit the problem of -gram extraction in the differential privacy setting. In this problem, given a corpus of private text data, the goal is to release as many -grams as…

cs.DS2021

Numerical Composition of Differential Privacy

Sivakanth Gopi, Yin Tat Lee, Lukas Wutschitz

We give a fast algorithm to optimally compose privacy guarantees of differentially private (DP) algorithms to arbitrary accuracy. Our method is based on the notion of privacy loss…

cs.LG20218 cited

Fast and Memory Efficient Differentially Private-SGD via JL Projections

Zhiqi Bu, Sivakanth Gopi, Janardhan Kulkarni +3

Differentially Private-SGD (DP-SGD) of Abadi et al. (2016) and its variations are the only known algorithms for private training of large scale neural networks. This algorithm requ…

cs.DS2020

Locally Private Hypothesis Selection

Sivakanth Gopi, Gautam Kamath, Janardhan Kulkarni +3

We initiate the study of hypothesis selection under local differential privacy. Given samples from an unknown probability distribution and a set of probability distribution…

cs.CC2019

CSPs with Global Modular Constraints: Algorithms and Hardness via Polynomial Representations

Joshua Brakensiek, Sivakanth Gopi, Venkatesan Guruswami

We study the complexity of Boolean constraint satisfaction problems (CSPs) when the assignment must have Hamming weight in some congruence class modulo M, for various choices of th…

cs.CC2018

Spanoids - an abstraction of spanning structures, and a barrier for LCCs

Zeev Dvir, Sivakanth Gopi, Yuzhou Gu +1

We introduce a simple logical inference structure we call a (generalizing the notion of a matroid), which captures well-studied problems in several areas. These…