Výsledky vyhledávání - "Mitsuji Muneyasu"

Akademický článek

ConfidentMix: Confidence-Guided Mixup for Learning With Noisy Labels

Autor: Ryota Higashimoto, Soh Yoshida, Mitsuji Muneyasu

Publikováno v: IEEE Access, Vol 12, Pp 58519-58531 (2024)

Deep neural networks (DNNs) have proven highly effective in various computational tasks, but their success depends largely on access to large datasets with accurate labels. Obtaining such data may be challenging and costly in real-world scenarios. Co

Externí odkaz: https://doaj.org/article/aa2d6fc03a194480a46540accadfcdf2

Zobrazit plný text záznamu

Akademický článek

Graph-Based Interpretability for Fake News Detection through Topic- and Propagation-Aware Visualization

Autor: Kayato Soga, Soh Yoshida, Mitsuji Muneyasu

Publikováno v: Computation, Vol 12, Iss 4, p 82 (2024)

In the context of the increasing spread of misinformation via social network services, in this study, we addressed the critical challenge of detecting and explaining the spread of fake news. Early detection methods focused on content analysis, wherea

Externí odkaz: https://doaj.org/article/b24db7a313b44125ace8bda3fffc18d2

Zobrazit plný text záznamu

Akademický článek

CRAS: Curriculum Regularization and Adaptive Semi-Supervised Learning with Noisy Labels

Autor: Ryota Higashimoto, Soh Yoshida, Mitsuji Muneyasu

Publikováno v: Applied Sciences, Vol 14, Iss 3, p 1208 (2024)

This paper addresses the performance degradation of deep neural networks caused by learning with noisy labels. Recent research on this topic has exploited the memorization effect: networks fit data with clean labels during the early stages of learnin

Externí odkaz: https://doaj.org/article/eb842868c4534db9a1fe829bcbf67088

Zobrazit plný text záznamu

Akademický článek

Training Robust Deep Neural Networks on Noisy Labels Using Adaptive Sample Selection With Disagreement

Autor: Hiroshi Takeda, Soh Yoshida, Mitsuji Muneyasu

Publikováno v: IEEE Access, Vol 9, Pp 141131-141143 (2021)

Learning with noisy labels is one of the most practical but challenging tasks in deep learning. One promising way to treat noisy labels is to use the small-loss trick based on the memorization effect, that is, clean and noisy samples are identified b

Externí odkaz: https://doaj.org/article/1576c61166b146aa925b8065e95c97bf

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání