H. Brendan McMahan

Training Production Language Models without Memorizing User Data

Sep 21, 2020
Swaroop Ramaswamy, Om Thakkar, Rajiv Mathews, Galen Andrew, H. Brendan McMahan, Françoise Beaufays

Access Paper or Ask Questions

Privacy Amplification via Random Check-Ins

Jul 30, 2020
Borja Balle, Peter Kairouz, H. Brendan McMahan, Om Thakkar, Abhradeep Thakurta

* Updated proof for $(\epsilon_0, \delta_0)$-DP local randomizers

Access Paper or Ask Questions

Adaptive Federated Optimization

Feb 29, 2020
Sashank Reddi, Zachary Charles, Manzil Zaheer, Zachary Garrett, Keith Rush, Jakub Konečný, Sanjiv Kumar, H. Brendan McMahan

Access Paper or Ask Questions

Is Local SGD Better than Minibatch SGD?

Feb 18, 2020
Blake Woodworth, Kumar Kshitij Patel, Sebastian U. Stich, Zhen Dai, Brian Bullins, H. Brendan McMahan, Ohad Shamir, Nathan Srebro

* 29 pages

Access Paper or Ask Questions

Advances and Open Problems in Federated Learning

Dec 10, 2019
Peter Kairouz, H. Brendan McMahan, Brendan Avent, Aurélien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Keith Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, Rafael G. L. D'Oliveira, Salim El Rouayheb, David Evans, Josh Gardner, Zachary Garrett, Adrià Gascón, Badih Ghazi, Phillip B. Gibbons, Marco Gruteser, Zaid Harchaoui, Chaoyang He, Lie He, Zhouyuan Huo, Ben Hutchinson, Justin Hsu, Martin Jaggi, Tara Javidi, Gauri Joshi, Mikhail Khodak, Jakub Konečný, Aleksandra Korolova, Farinaz Koushanfar, Sanmi Koyejo, Tancrède Lepoint, Yang Liu, Prateek Mittal, Mehryar Mohri, Richard Nock, Ayfer Özgür, Rasmus Pagh, Mariana Raykova, Hang Qi, Daniel Ramage, Ramesh Raskar, Dawn Song, Weikang Song, Sebastian U. Stich, Ziteng Sun, Ananda Theertha Suresh, Florian Tramèr, Praneeth Vepakomma, Jianyu Wang, Li Xiong, Zheng Xu, Qiang Yang, Felix X. Yu, Han Yu, Sen Zhao

Access Paper or Ask Questions

Can You Really Backdoor Federated Learning?

Dec 02, 2019
Ziteng Sun, Peter Kairouz, Ananda Theertha Suresh, H. Brendan McMahan

* To appear at the 2nd International Workshop on Federated Learning for Data Privacy and Confidentiality at NeurIPS 2019

Access Paper or Ask Questions

Generative Models for Effective ML on Private, Decentralized Datasets

Nov 15, 2019
Sean Augenstein, H. Brendan McMahan, Daniel Ramage, Swaroop Ramaswamy, Peter Kairouz, Mingqing Chen, Rajiv Mathews, Blaise Aguera y Arcas

* 27 pages, 8 figures

Access Paper or Ask Questions

Expanding the Reach of Federated Learning by Reducing Client Resource Requirements

Jan 08, 2019
Sebastian Caldas, Jakub Konečny, H. Brendan McMahan, Ameet Talwalkar

Access Paper or Ask Questions

A General Approach to Adding Differential Privacy to Iterative Training Procedures

Dec 15, 2018
H. Brendan McMahan, Galen Andrew

* Presented at NeurIPS 2018 workshop on Privacy Preserving Machine Learning

Access Paper or Ask Questions

Learning Differentially Private Recurrent Language Models

Feb 24, 2018
H. Brendan McMahan, Daniel Ramage, Kunal Talwar, Li Zhang

* Camera-ready ICLR 2018 version, minor edits from previous

Access Paper or Ask Questions

Federated Learning: Strategies for Improving Communication Efficiency

Oct 30, 2017
Jakub Konečný, H. Brendan McMahan, Felix X. Yu, Peter Richtárik, Ananda Theertha Suresh, Dave Bacon

Access Paper or Ask Questions

Distributed Mean Estimation with Limited Communication

Sep 25, 2017
Ananda Theertha Suresh, Felix X. Yu, Sanjiv Kumar, H. Brendan McMahan

Access Paper or Ask Questions

On the Protection of Private Information in Machine Learning Systems: Two Recent Approaches

Aug 26, 2017
Martín Abadi, Úlfar Erlingsson, Ian Goodfellow, H. Brendan McMahan, Ilya Mironov, Nicolas Papernot, Kunal Talwar, Li Zhang

* IEEE 30th Computer Security Foundations Symposium (CSF), pages 1--6, 2017

Access Paper or Ask Questions

Communication-Efficient Learning of Deep Networks from Decentralized Data

Feb 28, 2017
H. Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, Blaise Agüera y Arcas

* Proceedings of the 20 th International Conference on Artificial Intelligence and Statistics (AISTATS) 2017. JMLR: W&CP volume 54
* This version updates the large-scale LSTM experiments, along with other minor changes. In earlier versions, an inconsistency in our implementation of FedSGD caused us to report much lower learning rates for the large-scale LSTM. We reran these experiments, and also found that fewer local epochs offers better performance, leading to slightly better results for FedAvg than previously reported

Access Paper or Ask Questions

H. Brendan McMahan

Training Production Language Models without Memorizing User Data

Privacy Amplification via Random Check-Ins

Adaptive Federated Optimization

Is Local SGD Better than Minibatch SGD?

Advances and Open Problems in Federated Learning

Can You Really Backdoor Federated Learning?

Generative Models for Effective ML on Private, Decentralized Datasets

Differentially Private Learning with Adaptive Clipping

SysML: The New Frontier of Machine Learning Systems

Semi-Cyclic Stochastic Gradient Descent

Towards Federated Learning at Scale: System Design

LEAF: A Benchmark for Federated Settings

Expanding the Reach of Federated Learning by Reducing Client Resource Requirements

A General Approach to Adding Differential Privacy to Iterative Training Procedures

Learning Differentially Private Recurrent Language Models

Federated Learning: Strategies for Improving Communication Efficiency

Distributed Mean Estimation with Limited Communication

On the Protection of Private Information in Machine Learning Systems: Two Recent Approaches

Communication-Efficient Learning of Deep Networks from Decentralized Data

Practical Secure Aggregation for Federated Learning on User-Held Data

Deep Learning with Differential Privacy

Federated Optimization: Distributed Machine Learning for On-Device Intelligence

A Survey of Algorithms and Analysis for Adaptive Online Learning

Unconstrained Online Linear Learning in Hilbert Spaces: Minimax Algorithms and Normal Approximations

Large-Scale Learning with Less RAM via Randomization