Using Machine Learning to Quantify the Robustness of Network Controllability

Using Machine Learning to Quantify the Robustness of Network Controllability

Dhiman, Ashish (Student TU Delft)
Sun, P. (TU Delft Network Architectures and Services)
Kooij, Robert (TU Delft Network Architectures and Services)

Renault, Éric (editor)
Boumerdassi, Selma (editor)
Mühlethaler, Paul (editor)

2021

This paper presents machine learning based approximations for the minimum number of driver nodes needed for structural controllability of networks under link-based random and targeted attacks. We compare our approximations with existing analytical approximations and show that our machine learning based approximations significantly outperform the existing closed-form analytical approximations in case of both synthetic and real-world networks. Apart from targeted attacks based upon the removal of so-called critical links, we also propose analytical approximations for out-in degree-based attacks.

Driver nodes
Machine learning
Network controllability
Network robustness

http://resolver.tudelft.nl/uuid:456d53df-8a2c-41c8-86c9-4ba4a50732c7

https://doi.org/10.1007/978-3-030-70866-5_2

Springer

2021-11-29

9783030708658

Machine Learning for Networking - Third International Conference, MLN 2020, Revised Selected Papers

3rd International Conference on Machine Learning for Networking, MLN 2020, 2020-11-24 → 2020-11-26, Paris, France

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 0302-9743, 12629 LNCS

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

conference paper

© 2021 Ashish Dhiman, P. Sun, Robert Kooij