paper-with-me

홈 › Papers

How reparametrization trick broke differentially-private text representation learning

2022-02-24 · ACL 2022 5 · Ivan Habernal

As privacy gains traction in the NLP community, researchers have started adopting various approaches to privacy-preserving methods. One of the favorite privacy frameworks, differential privacy (DP), is perhaps the most compelling thanks to its fundamental theoretical guarantees. Despite the apparent simplicity of the general concept of differential privacy, it seems non-trivial to get it right when applying it to NLP. In this short paper, we formally analyze several recent NLP papers proposing text representation learning using DPText (Beigi et al., 2019a,b; Alnasser et al., 2021; Beigi et al., 2021) and reveal their false claims of being differentially private. Furthermore, we also show a simple yet general empirical sanity check to determine whether a given implementation of a DP mechanism almost certainly violates the privacy loss guarantees. Our main goal is to raise awareness and help the community understand potential pitfalls of applying differential privacy to text representation learning.

📄 PDF Abstract BibTeX arXiv:2202.12138

Code (1)

trusthlt/acl2022-reparametrization-trick-broke-differential-privacy 공식 구현

Tasks

Privacy PreservingRepresentation Learning

Similar Papers 제목 키워드 기반

Large Scale Private Learning via Low-rank Reparametrization

2021-06-17 · Da Yu, Huishuai Zhang, Wei Chen, Jian Yin 외

We propose a reparametrization scheme to address the challenges of applying differentially private SGD on large neural networks, which are 1) the huge memory cost of storing individual gradients, 2) the added noise suffe…

A Scalable Laplace Approximation for Neural Networks

2018-01-01 · ICLR 2018 1 · Hippolyt Ritter, Aleksandar Botev, David Barber

Pytorch implementations of Bayes By Backprop, MC Dropout, SGLD, the Local Reparametrization Trick, KF-Laplace and more

Bayesian Inference

Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks

2015-12-23 · Chunyuan Li, Changyou Chen, David Carlson, Lawrence Carin

Pytorch implementations of Bayes By Backprop, MC Dropout, SGLD, the Local Reparametrization Trick, KF-Laplace and more

spred: Solving $L_1$ Penalty with SGD

2022-10-03 · Liu Ziyin, ZiHao Wang

We propose to minimize a generic differentiable objective with $L_1$ constraint using a simple reparametrization and straightforward stochastic gradient descent. Our proposal is the direct generalization of previous idea…

Inductive BiasNeural Network Compression

An Empirical Analysis of Measure-Valued Derivatives for Policy Gradients

2021-07-20 · João Carvalho, Davide Tateo, Fabio Muratore, Jan Peters

Reinforcement learning methods for robotics are increasingly successful due to the constant development of better policy gradient techniques. A precise (low variance) and accurate (low bias) gradient estimator is crucial…