Multiple Instance Learning With Instance-Level Positive-Unlabeled Learning in Anomaly Detection

We propose a method for learning a classifier that accurately predicts both instance and bag classes in multiple instance learning (MIL) for anomaly detection, achieving significant performance improvement. MIL, a form of weakly supervised learning, represents datasets as sets of bags labeled as eit...

Descrizione completa

Dettagli Bibliografici

Autori principali:	Ryosuke Matsuo, Shinya Yasuda, Hiroshi Yoshida
Natura:	Artigo
Lingua:	Inglês
Pubblicazione:	IEEE 2025-01-01
Serie:	IEEE Access
Soggetti:	Anomaly detection multiple instance learning positive and unlabeled learning weakly supervised learning
Accesso online:	https://ieeexplore.ieee.org/document/11030552/
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!