Schynol, Lukas ; Pesavento, Marius (2023)
Deep Unrolling for Anomaly Detection in Network Flows.
9th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing. Herradura, Costa Rica (10.12.2023-13.12.2023)
doi: 10.1109/CAMSAP58249.2023.10403513
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
Anomaly detection becomes increasingly important in the design of future resilient communication systems. In this work, anomaly detection in network flows with incomplete measurements on the basis of robust PCA, where normal flows are characterized as low-rank components and anomalies as sparse components, is considered. Based on the block-successive convex approximation framework, we first introduce a novel model-based algorithm for normal and anomalous traffic recovery. Since robust-PCA-based anomaly detection alone is suboptimal in terms of the receiver operating characteristic, we apply deep unrolling to this algorithm and use a homotopy optimization method to train the resulting deep network architecture to explicitly optimize the area under the curve of the receiver oper-ating characteristic. Thereby, the domain knowledge introduced by robust PCA is retained. Our deep unrolling-based network architecture outperforms the classical methods while generalizing well and featuring excellent data efficiency.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2023 |
| Autor(en): | Schynol, Lukas ; Pesavento, Marius |
| Art des Eintrags: | Bibliographie |
| Titel: | Deep Unrolling for Anomaly Detection in Network Flows |
| Sprache: | Deutsch |
| Publikationsjahr: | 14 Dezember 2023 |
| Verlag: | IEEE |
| Buchtitel: | 2023 IEEE 9th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) |
| Veranstaltungstitel: | 9th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing |
| Veranstaltungsort: | Herradura, Costa Rica |
| Veranstaltungsdatum: | 10.12.2023-13.12.2023 |
| DOI: | 10.1109/CAMSAP58249.2023.10403513 |
| Kurzbeschreibung (Abstract): | Anomaly detection becomes increasingly important in the design of future resilient communication systems. In this work, anomaly detection in network flows with incomplete measurements on the basis of robust PCA, where normal flows are characterized as low-rank components and anomalies as sparse components, is considered. Based on the block-successive convex approximation framework, we first introduce a novel model-based algorithm for normal and anomalous traffic recovery. Since robust-PCA-based anomaly detection alone is suboptimal in terms of the receiver operating characteristic, we apply deep unrolling to this algorithm and use a homotopy optimization method to train the resulting deep network architecture to explicitly optimize the area under the curve of the receiver oper-ating characteristic. Thereby, the domain knowledge introduced by robust PCA is retained. Our deep unrolling-based network architecture outperforms the classical methods while generalizing well and featuring excellent data efficiency. |
| Freie Schlagworte: | Conferences, Optimization methods, Receivers, Network architecture, Approximation algorithms, Anomaly detection, Principal component analysis, Deep unrolling, anomaly detection |
| Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Nachrichtentechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Nachrichtentechnik > Nachrichtentechnische Systeme |
| Hinterlegungsdatum: | 16 Jan 2025 12:50 |
| Letzte Änderung: | 16 Jan 2025 12:50 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum