Protocon: Pseudo-label refinement via online clustering and prototypical consistency for efficient semi-supervised learning

I Nassar, M Hayat, E Abbasnejad… - Proceedings of the …, 2023 - openaccess.thecvf.com
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …
被引用次数：15 相关文章

ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-supervised Learning

I Nassar, M Hayat, E Abbasnejad, H Rezatofighi… - arXiv preprint arXiv …, 2023 - arxiv.org
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …

Protocon: Pseudo-Label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-Supervised Learning

I Nassar, M Hayat, E Abbasnejad… - 2023 IEEE/CVF …, 2023 - computer.org
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …

Protocon: Pseudo-Label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-Supervised Learning

I Nassar, M Hayat, E Abbasnejad… - 2023 IEEE/CVF …, 2023 - ieeexplore.ieee.org
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …

Protocon: pseudo-label refinement via online clustering and prototypical consistency for efficient semi-supervised learning

I Nassar, M Hayat, E Abbasnejad… - … on Computer Vision …, 2023 - research.monash.edu
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …

ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-supervised Learning

I Nassar, M Hayat, E Abbasnejad… - arXiv e …, 2023 - ui.adsabs.harvard.edu
Confidence-based pseudo-labeling is among the dominant approaches in semi-supervised
learning (SSL). It relies on including high-confidence predictions made on unlabeled data as …