Applications of Self-Supervised Learning to Biomedical Signals: A Survey

Pup, Federico Del; Atzori, Manfredo

doi:10.1109/access.2023.3344531

Cited by 4 publications

(2 citation statements)

References 185 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Del Pup conducted a comprehensive survey on the applications of SSL to biomedical signals, highlighting its potential for extracting meaningful representations from diverse biomedical data sources [46]. This is particularly relevant considering the inherent challenge faced by both the biomedical and cybersecurity domains in acquiring labeled data at scale, owing to factors such as privacy concerns, data scarcity, and the need for domain expertise.…”

Section: B Self-supervised Learningmentioning

confidence: 99%

MalSSL—Self-Supervised Learning for Accurate and Label-Efficient Malware Classification

Ismail,

Hendrawan,

Rahardjo

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Malware classification with supervised learning requires a large dataset which needs an expensive and time-consuming labeling process. In this paper, we explore the efficacy of self-supervised learning techniques for malware classification. We propose MalSSL, a self-supervised learning-based method utilizing image representation to classify malware. MalSSL classifies unlabeled malware images using contrastive learning and data augmentation. The model is initially trained on an unlabeled Imagenette dataset as a pretext task and subsequently retrained on an unlabeled malware dataset in downstream tasks. Two downstream tasks were employed to evaluate the system: 1) Malware family classification and 2) Malware Benign classification. The obtained results include an accuracy of 98.4 % on the malware family classification experiment on the Malimg dataset and an accuracy of 96.2 % on the malware and benign dataset (Maldeb dataset). Our findings suggest that the proposed system accurately classifies malware without the need for labeled data, displaying higher accuracy compared to other self-supervised methods. This research not only contributes to advancing the state-of-the-art in malware classification but also underscores the potential of self-supervised learning methods as a viable solution for addressing the dynamic landscape of malware threats.

show abstract

Section: B Self-supervised Learningmentioning

confidence: 99%

MalSSL—Self-Supervised Learning for Accurate and Label-Efficient Malware Classification

Ismail,

Hendrawan,

Rahardjo

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Expectationmaximization (EM) and multiple instance learning jointly infer labels and recognize discriminative regions, enabling training from cheaper forms of weak supervision. G. Self-supervised/unsupervised learning Self-supervised learning has also gained vast attention in CV by enabling pre-training from sheer ubiquity of unlabeled visual data [384]- [393]. Pretext tasks like predicting image rotations, solving jigsaw puzzles, or counting pixel colors allow models to learn rich visual representations applicable to downstream tasks.…”

Section: F Weakly-supervised Learningmentioning

confidence: 99%