Zobrazeno 1 - 10
of 61
pro vyhledávání: '"Asoodeh, Shahab"'
Local differential privacy (LDP) is increasingly employed in privacy-preserving machine learning to protect user data before sharing it with an untrusted aggregator. Most LDP methods assume that users possess only a single data record, which is a sig
Externí odkaz:
http://arxiv.org/abs/2411.08791
The sampling problem under local differential privacy has recently been studied with potential applications to generative models, but a fundamental analysis of its privacy-utility trade-off (PUT) remains incomplete. In this work, we define the fundam
Externí odkaz:
http://arxiv.org/abs/2410.22699
Autor:
Ghoukasian, Hrad, Asoodeh, Shahab
In this work, we investigate binary classification under the constraints of both differential privacy and fairness. We first propose an algorithm based on the decoupling technique for learning a classifier with only fairness guarantee. This algorithm
Externí odkaz:
http://arxiv.org/abs/2402.15603
We study the problem of hypothesis selection under the constraint of local differential privacy. Given a class $\mathcal{F}$ of $k$ distributions and a set of i.i.d. samples from an unknown distribution $h$, the goal of hypothesis selection is to pic
Externí odkaz:
http://arxiv.org/abs/2312.05645
Autor:
Asoodeh, Shahab, Diaz, Mario
The Noisy-SGD algorithm is widely used for privately training machine learning models. Traditional privacy analyses of this algorithm assume that the internal state is publicly revealed, resulting in privacy loss bounds that increase indefinitely wit
Externí odkaz:
http://arxiv.org/abs/2305.09903
Publikováno v:
2023 IEEE International Symposium on Information Theory (ISIT), Taipei, Taiwan, 2023, pp. 1478-1483
The information bottleneck (IB) method aims to find compressed representations of a variable $X$ that retain the most relevant information about a target variable $Y$. We show that for a wide family of distributions -- namely, when $Y$ is generated b
Externí odkaz:
http://arxiv.org/abs/2305.07000
Autor:
Asoodeh, Shahab, Zhang, Huanyu
We investigate the contraction properties of locally differentially private mechanisms. More specifically, we derive tight upper bounds on the divergence between $PK$ and $QK$ output distributions of an $\epsilon$-LDP mechanism $K$ in terms of a dive
Externí odkaz:
http://arxiv.org/abs/2210.13386
Autor:
Alghamdi, Wael, Asoodeh, Shahab, Calmon, Flavio P., Gomez, Juan Felipe, Kosut, Oliver, Sankar, Lalitha, Wei, Fei
We introduce a new differential privacy (DP) accountant called the saddle-point accountant (SPA). SPA approximates privacy guarantees for the composition of DP mechanisms in an accurate and fast manner. Our approach is inspired by the saddle-point me
Externí odkaz:
http://arxiv.org/abs/2208.09595
Most differential privacy mechanisms are applied (i.e., composed) numerous times on sensitive data. We study the design of optimal differential privacy mechanisms in the limit of a large number of compositions. As a consequence of the law of large nu
Externí odkaz:
http://arxiv.org/abs/2207.00420
Autor:
Alghamdi, Wael, Hsu, Hsiang, Jeong, Haewon, Wang, Hao, Michalak, P. Winston, Asoodeh, Shahab, Calmon, Flavio P.
We consider the problem of producing fair probabilistic classifiers for multi-class classification tasks. We formulate this problem in terms of "projecting" a pre-trained (and potentially unfair) classifier onto the set of models that satisfy target
Externí odkaz:
http://arxiv.org/abs/2206.07801