SimCDL: A Simple Framework for Contrastive Dictionary Learning
Applied Sciences · 05.11.2024 · 14 (22) · 10082
DOI: 10.3390/app142210082 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 05.11.2024. Verfasst von Denis C. Ilie-Ablachim und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Research, Innovation and Digitization, CNCS—UEFISCDI.
Abstract (Originalsprache)
In this paper, we propose a novel approach to the dictionary learning (DL) initialization problem, leveraging the SimCLR framework from deep learning in a self-supervised manner. Dictionary learning seeks to represent signals as sparse combinations of dictionary atoms, but effective initialization remains challenging. By applying contrastive learning, we encourage similar representations for augmented versions of the same sample while distinguishing between different samples. This results in a more diverse and incoherent set of atoms, which enhances the performance of DL applications in classification and anomaly detection tasks. Our experiments across several benchmark datasets demonstrate the effectiveness of our method for improving dictionary learning initialization and its subsequent impact on performance in various applications.
Förderung
- Ministry of Research, Innovation and Digitization, CNCS—UEFISCDI
Zitieren
APA
Ilie-Ablachim, D. C., & Dumitrescu, B. (2024). SimCDL: A Simple Framework for Contrastive Dictionary Learning. Applied Sciences, 14(22), 10082. https://doi.org/10.3390/app142210082
Vancouver
Ilie-Ablachim DC, Dumitrescu B. SimCDL: A Simple Framework for Contrastive Dictionary Learning. Applied Sciences. 2024;14(22):10082. doi:10.3390/app142210082
BibTeX
@article{IlieAblachim2024,
author = {Ilie-Ablachim, Denis C. and Dumitrescu, Bogdan},
title = {SimCDL: A Simple Framework for Contrastive Dictionary Learning},
journal = {Applied Sciences},
year = {2024},
volume = {14},
number = {22},
pages = {10082},
doi = {10.3390/app142210082}
}
EUVIMED