Differentially Private Federated Learning: A client-level perspective

Introduction:

Federated Learning is a privacy preserving decentralized learning protocol introduced my Google. Multiple clients jointly learn a model without data centralization. Centralization is pushed from data space to parameter space: https://research.google.com/pubs/pub44822.html [1]. Differential privacy in deep learning is concerned with preserving privacy of individual data points: https://arxiv.org/abs/1607.00133 [2]. In this work we combine the notion of both by making federated learning differentially private. We focus on preserving privacy for the entire data set of a client. For more information, please refer to: https://arxiv.org/abs/1712.07557v2.

This code simulates a federated setting and enables federated learning with differential privacy. The privacy accountant used is from https://arxiv.org/abs/1607.00133 [2]. The files: accountant.py, utils.py, gaussian_moments.py are taken from: https://github.com/tensorflow/models/tree/master/research/differential_privacy

Note that the privacy agent is not completely set up yet (especially for more than 100 clients). It has to be specified manually or otherwise parameters 'm' and 'sigma' need to be specified.

Installation

1- If not already done, install Tensorflow

2- Downlad the directory this README is part of

3- If using macOS, simply run:

bash RUNME.sh

to download the MNIST data-sets, create clients and getting started. For more information on the individual functions, please refer to their doc strings.

How to Cite

If you use this code or the pretrained models in your research, please cite:

@ARTICLE{2017arXiv171207557G,
   author = {{Geyer}, R.~C. and {Klein}, T. and {Nabi}, M.},
    title = "{Differentially Private Federated Learning: A Client Level Perspective}",
  journal = {ArXiv e-prints},
archivePrefix = "arXiv",
   eprint = {1712.07557},
 primaryClass = "cs.CR",
 keywords = {Computer Science - Cryptography and Security, Computer Science - Learning, Statistics - Machine Learning},
     year = 2017,
    month = dec,
   adsurl = {http://adsabs.harvard.edu/abs/2017arXiv171207557G},
  adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

Reference(s)

[1] H. Brendan McMahan et al., Communication-Efficient Learning of Deep Networks from Decentralized Data, 2017, http://arxiv.org/abs/1602.05629.

[2] Martin Abadi et al., Deep Learning with Differential Privacy, 2016, https://arxiv.org/abs/1607.00133.