Publication
Adaptative Perturbation Patterns: Realistic Adversarial Learning for Robust Intrusion Detection
| dc.contributor.author | Vitorino, João | |
| dc.contributor.author | Oliveira, Nuno | |
| dc.contributor.author | Praça, Isabel | |
| dc.date.accessioned | 2023-01-25T11:37:47Z | |
| dc.date.available | 2023-01-25T11:37:47Z | |
| dc.date.issued | 2022-03-08 | |
| dc.description.abstract | Adversarial attacks pose a major threat to machine learning and to the systems that rely on it. In the cybersecurity domain, adversarial cyber-attack examples capable of evading detection are especially concerning. Nonetheless, an example generated for a domain with tabular data must be realistic within that domain. This work establishes the fundamental constraint levels required to achieve realism and introduces the Adaptative Perturbation Pattern Method (A2PM) to fulfill these constraints in a gray-box setting. A2PM relies on pattern sequences that are independently adapted to the characteristics of each class to create valid and coherent data perturbations. The proposed method was evaluated in a cybersecurity case study with two scenarios: Enterprise and Internet of Things (IoT) networks. Multilayer Perceptron (MLP) and Random Forest (RF) classifiers were created with regular and adversarial training, using the CIC-IDS2017 and IoT-23 datasets. In each scenario, targeted and untargeted attacks were performed against the classifiers, and the generated examples were compared with the original network traffic flows to assess their realism. The obtained results demonstrate that A2PM provides a scalable generation of realistic adversarial examples, which can be advantageous for both adversarial training and attacks. | pt_PT |
| dc.description.sponsorship | The present work has received funding from the European Union’s Horizon 2020 research and innovation program, under project SeCoIIA (grant agreement no. 871967). This work has also received funding from UIDP/00760/2020. | pt_PT |
| dc.description.version | info:eu-repo/semantics/publishedVersion | pt_PT |
| dc.identifier.doi | 10.3390/fi14040108 | pt_PT |
| dc.identifier.uri | http://hdl.handle.net/10400.22/21851 | |
| dc.language.iso | eng | pt_PT |
| dc.peerreviewed | yes | pt_PT |
| dc.publisher | MDPI | pt_PT |
| dc.relation | Secure Collaborative Intelligent Industrial Assets | |
| dc.relation | Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development | |
| dc.relation.publisherversion | https://www.mdpi.com/1999-5903/14/4/108 | pt_PT |
| dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/ | pt_PT |
| dc.subject | Realistic adversarial examples | pt_PT |
| dc.subject | Adversarial attacks | pt_PT |
| dc.subject | Adversarial robustness | pt_PT |
| dc.subject | Machine learning | pt_PT |
| dc.subject | Tabular data | pt_PT |
| dc.subject | Intrusion detection | pt_PT |
| dc.title | Adaptative Perturbation Patterns: Realistic Adversarial Learning for Robust Intrusion Detection | pt_PT |
| dc.type | journal article | |
| dspace.entity.type | Publication | |
| oaire.awardTitle | Secure Collaborative Intelligent Industrial Assets | |
| oaire.awardTitle | Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development | |
| oaire.awardURI | info:eu-repo/grantAgreement/EC/H2020/871967/EU | |
| oaire.awardURI | info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F00760%2F2020/PT | |
| oaire.citation.issue | 4 | pt_PT |
| oaire.citation.startPage | 108 | pt_PT |
| oaire.citation.title | Future Internet | pt_PT |
| oaire.citation.volume | 14 | pt_PT |
| oaire.fundingStream | H2020 | |
| oaire.fundingStream | 6817 - DCRRNI ID | |
| person.familyName | Machado Vitorino | |
| person.familyName | Oliveira | |
| person.familyName | Praça | |
| person.givenName | João Pedro | |
| person.givenName | Nuno | |
| person.givenName | Isabel | |
| person.identifier | HlqCxhoAAAAJ | |
| person.identifier | 299522 | |
| person.identifier.ciencia-id | 3312-592F-B628 | |
| person.identifier.ciencia-id | 3E1B-B728-9524 | |
| person.identifier.ciencia-id | C710-4218-1BFF | |
| person.identifier.orcid | 0000-0002-4968-3653 | |
| person.identifier.orcid | 0000-0002-5030-7751 | |
| person.identifier.orcid | 0000-0002-2519-9859 | |
| person.identifier.rid | K-8430-2014 | |
| person.identifier.scopus-author-id | 57579914600 | |
| person.identifier.scopus-author-id | 22734900800 | |
| project.funder.identifier | http://doi.org/10.13039/501100008530 | |
| project.funder.identifier | http://doi.org/10.13039/501100001871 | |
| project.funder.name | European Commission | |
| project.funder.name | Fundação para a Ciência e a Tecnologia | |
| rcaap.rights | openAccess | pt_PT |
| rcaap.type | article | pt_PT |
| relation.isAuthorOfPublication | 4435bb64-8a77-407b-ba18-4833d26e72ae | |
| relation.isAuthorOfPublication | e49f38bc-accb-44eb-8f49-7e7e555f34a5 | |
| relation.isAuthorOfPublication | ee4ecacd-c6c6-41e8-bca1-21a60ff05f50 | |
| relation.isAuthorOfPublication.latestForDiscovery | ee4ecacd-c6c6-41e8-bca1-21a60ff05f50 | |
| relation.isProjectOfPublication | 50b4a014-9a9b-44b6-ae09-72173c0acae9 | |
| relation.isProjectOfPublication | 6eb94c83-adf9-4d9d-a75c-be95f44e3ca5 | |
| relation.isProjectOfPublication.latestForDiscovery | 6eb94c83-adf9-4d9d-a75c-be95f44e3ca5 |