Metraipolis › archive › dc_a9bf188b1fee
LiNC: Lightweight Noise Correction via Per-Sample Trust and Gaussian Mixture Modeling — arXiv:2608.04147
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04147
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:13:35.073Z
- SHA-256 of the archived text
e1f45aa0513459561de48e939022c1e283e6890a9940f78add76b2d34877712e- Stable URL of this page
https://metraipolis.com/documento/dc_a9bf188b1fee
How to cite
LiNC: Lightweight Noise Correction via Per-Sample Trust and Gaussian Mixture Modeling — arXiv:2608.04147 — arXiv (en), https://arxiv.org/abs/2608.04147 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a9bf188b1fee (Metraipolis archive dc_a9bf188b1fee, sha256 e1f45aa0513459561de48e939022c1e283e6890a9940f78add76b2d34877712e)
BibTeX
@misc{dca9bf188b1fee,
title = {LiNC: Lightweight Noise Correction via Per-Sample Trust and Gaussian Mixture Modeling — arXiv:2608.04147},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04147},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a9bf188b1fee; sha256 e1f45aa0513459561de48e939022c1e283e6890a9940f78add76b2d34877712e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04147. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Label noise is common in medical imaging datasets due to factors such as inter-rater variability, annotation errors, and ambiguous cases. This can severely undermine the reliability and clinical effectiveness of machine learning models trained using those datasets. To address this challenge, we introduce Lightweight Noise Correction (LiNC), which adds a single trainable trust parameter per training sample and learns when to use the observed label and when to defer to the model during a standard training loop. The key idea is to train using a convex combination of the observed label and the model's own predictive distribution, controlled by a per-sample trust parameter. We show that the gradient of this objective drives trust values in opposite directions for clean versus noisy samples in the early training phase, yielding separable trust distributions. We use a 3-component Gaussian Mixture Model over the trust values to separate them into clean, ambiguous, and noisy cases and then execute a short soft-correction phase on the noisy cases and a final hard correction phase. Experiments on ten 2D datasets from MedMNISTv2 under label noise of up to 50% show consistent gains in accuracy and strong mislabel detection. LiNC adds negligible asymptotic overhead: the training-time complexity remains dominated by the base network, with additional memory growing linearly with the size of the training set. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.04147 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T11:02:13.483Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).