Pith. sign in

REVIEW

Machine learning prediction of network dynamics with privacy protection

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2206.04828 v1 pith:DUHSNX4T submitted 2022-06-10 physics.soc-ph

Machine learning prediction of network dynamics with privacy protection

classification physics.soc-ph
keywords datanetworkdynamicsprivacydifferentpartialpartiespredict
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Predicting network dynamics based on data, a problem with broad applications, has been studied extensively in the past, but most existing approaches assume that the complete set of historical data from the whole network is available. This requirement presents a great challenge in applications, especially for large, distributed networks in the real world, where data collection is accomplished by many clients in a parallel fashion. Often, each client only has the time series data from a partial set of nodes and the client has access to only partial timestamps of the whole time series data and partial structure of the network. Due to privacy concerns or license related issues, the data collected by different clients cannot be shared. To accurately predict the network dynamics while protecting the privacy of different parties is a critical problem in the modern time. Here, we propose a solution based on federated graph neural networks (FGNNs) that enables the training of a global dynamic model for all parties without data sharing. We validate the working of our FGNN framework through two types of simulations to predict a variety of network dynamics (four discrete and three continuous dynamics). As a significant real-world application, we demonstrate successful prediction of State-wise influenza spreading in the USA. Our FGNN scheme represents a general framework to predict diverse network dynamics through collaborative fusing of the data from different parties without disclosing their privacy.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.