Loss Functions for Clustering in Multi-instance Learning

0533916 - ÚI 2021 RIV DE eng C - Conference Paper (international conference)
Dědič, M. - Pevný, T. - Bajer, L. - Holeňa, Martin
Loss Functions for Clustering in Multi-instance Learning.
Proceedings of the 20th Conference Information Technologies - Applications and Theory. Aachen: Technical University & CreateSpace Independent Publishing, 2020 - (Holeňa, M.; Horváth, T.; Kelemenová, A.; Mráz, F.; Pardubská, D.; Plátek, M.; Sosík, P.), s. 137-146. CEUR Workshop Proceedings, 2718. ISSN 1613-0073.
[ITAT 2020: Information Technologies - Applications and Theory /20./. Oravská Lesná (SK), 18.09.2020-22.09.2020]
R&D Projects: GA ČR(CZ) GA18-18080S
Grant - others:GA ČR(CZ) GA18-21409S
Program: GA
Institutional support: RVO:67985807
Keywords : Representation learning * Multi-instance learning * Multi-instance clustering * Clustering loss functions * Intrusion detection
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
http://ceur-ws.org/Vol-2718/paper05.pdf

Multi-instance learning belongs to one of recently fast developing areas of machine learning. It is a supervised learning method and this paper reports research into its unsupervised counterpart, multi-instance clustering. Whereas traditional clustering clusters points, multiinstance clustering clusters bags, i.e. multisets of points or of other kinds of objects. The paper focuses on the problem of loss functions for clustering. Three sophisticated loss functions used for clustering of points, contrastive predictive coding, triplet loss and magnet loss, are elaborated for multi-instance clustering. Finally, they are compared on 18 benchmark datasets, as well as on a real-world dataset.
Permanent Link: http://hdl.handle.net/11104/0312145