# Local Graph-homomorphic Processing for Privatized Distributed Systems

arXiv: https://arxiv.org/abs/2210.15414 · HTML: https://mlsanity.com/paper/2210.15414/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 32 references that are in the map.
- First posted on arXiv 26 Oct 2022.

## Builds on (most cited first, 6 of 6)

- [A Survey on Distributed Machine Learning](https://mlsanity.com/paper/1912.09789/) (2019)
- [Differentially Private Distributed Online Learning](https://mlsanity.com/paper/1505.06556/) (2015)
- [A Graph Federated Architecture with Privacy Preserving Learning](https://mlsanity.com/paper/2104.13215/) (2021)
- [Scalable Privacy-Preserving Distributed Learning](https://mlsanity.com/paper/2005.09532/) (2020)
- [Graph-Homomorphic Perturbations for Private Decentralized Learning](https://mlsanity.com/paper/2010.12288/) (2020)
- [DFINITY Technology Overview Series, Consensus System](https://mlsanity.com/paper/1805.04548/) (2018)

## Built on by (most cited first, 1 of 1)

- [Dynamic Privacy Allocation for Locally Differentially Private Federated Learning with Composite Objectives](https://mlsanity.com/paper/2308.01139/) (2023)

## Abstract (arXiv)

We study the generation of dependent random numbers in a distributed fashion in order to enable privatized distributed learning by networked agents. We propose a method that we refer to as local graph-homomorphic processing; it relies on the construction of particular noises over the edges to ensure a certain level of differential privacy. We show that the added noise does not affect the performance of the learned model. This is a significant improvement to previous works on differential privacy for distributed algorithms, where the noise was added in a less structured manner without respecting the graph topology and has often led to performance deterioration. We illustrate the theoretical results by considering a linear regression problem over a network of agents.