paper-with-me

홈 › Papers

Whisper D-SGD: Correlated Noise Across Agents for Differentially Private Decentralized Learning

2025-01-24 · Angelo Rodio, Zheng Chen, Erik G. Larsson

Decentralized learning enables distributed agents to train a shared machine learning model through local computation and peer-to-peer communication. Although each agent retains its dataset locally, the communication of local models can still expose private information to adversaries. To mitigate these threats, local differential privacy (LDP) injects independent noise per agent, but it suffers a larger utility gap than central differential privacy (CDP). We introduce Whisper D-SGD, a novel covariance-based approach that generates correlated privacy noise across agents, unifying several state-of-the-art methods as special cases. By leveraging network topology and mixing weights, Whisper D-SGD optimizes the noise covariance to achieve network-wide noise cancellation. Experimental results show that Whisper D-SGD cancels more noise than existing pairwise-correlation schemes, substantially narrowing the CDP-LDP gap and improving model performance under the same privacy guarantees.

📄 PDF Abstract BibTeX arXiv:2501.14644

Code (1)

arodio/whisperdsgd 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Locally Differentially Private Online Federated Learning With Correlated Noise

2024-11-27 · Jiaojiao Zhang, Linglingzhi Zhu, Dominik Fay, Mikael Johansson

We introduce a locally differentially private (LDP) algorithm for online federated learning that employs temporally correlated noise to improve utility while preserving privacy. To address challenges posed by the correla…

Federated Learning

DP-λCGD: Efficient Noise Correlation for Differentially Private Model Training

2026-01-29 · Nikita P. Kalinin, Ryan McKenna, Rasmus Pagh, Christoph H. Lampert arxiv

Differentially private stochastic gradient descent (DP-SGD) is the gold standard for training machine learning models with formal differential privacy guarantees. Several recent extensions improve its accuracy by introdu…

AudioSAE: Towards Understanding of Audio-Processing Models with Sparse AutoEncoders

2026-02-04 · Georgii Aparin, Tasnima Sadekova, Alexey Rukhovich, Assel Yermekova 외 arxiv

Sparse Autoencoders (SAEs) are powerful tools for interpreting neural representations, yet their use in audio remains underexplored. We train SAEs across all encoder layers of Whisper and HuBERT, provide an extensive eva…

Distributed Differentially-Private Algorithms for Matrix and Tensor Factorization

2018-04-26 · Hafiz Imtiaz, Anand D. Sarwate

In many signal processing and machine learning applications, datasets containing private information are held at different locations, requiring the development of distributed privacy-preserving algorithms. Tensor and mat…

Privacy PreservingTensor Decomposition

Differentially Private Online Federated Learning with Correlated Noise

2024-03-25 · Jiaojiao Zhang, Linglingzhi Zhu, Mikael Johansson

We introduce a novel differentially private algorithm for online federated learning that employs temporally correlated noise to enhance utility while ensuring privacy of continuously released models. To address challenge…

Federated Learning