DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Youngdong | ko |
dc.contributor.author | Yun, Juseung | ko |
dc.contributor.author | Shon, Hyounguk | ko |
dc.contributor.author | Kim, Junmo | ko |
dc.date.accessioned | 2022-11-21T03:00:31Z | - |
dc.date.available | 2022-11-21T03:00:31Z | - |
dc.date.created | 2022-11-19 | - |
dc.date.issued | 2021-06 | - |
dc.identifier.citation | 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021, pp.9437 - 9446 | - |
dc.identifier.issn | 1063-6919 | - |
dc.identifier.uri | http://hdl.handle.net/10203/300213 | - |
dc.description.abstract | Training of Convolutional Neural Networks (CNNs) with data with noisy labels is known to be a challenge. Based on the fact that directly providing the label to the data (Positive Learning; PL) has a risk of allowing CNNs to memorize the contaminated labels for the case of noisy data, the indirect learning approach that uses complementary labels (Negative Learning for Noisy Labels; NLNL) has proven to be highly effective in preventing overfitting to noisy data as it reduces the risk of providing faulty target. NLNL further employs a three-stage pipeline to improve convergence. As a result, filtering noisy data through the NLNL pipeline is cumbersome, increasing the training cost. In this study, we propose a novel improvement of NLNL, named Joint Negative and Positive Learning (JNPL), that unifies the filtering pipeline into a single stage. JNPL trains CNN via two losses, NL+ and PL+, which are improved upon NL and PL loss functions, respectively. We analyze the fundamental issue of NL loss function and develop new NL+ loss function producing gradient that enhances the convergence of noisy data. Furthermore, PL+ loss function is designed to enable faster convergence to expected-to-be-clean data. We show that the NL+ and PL+ train CNN simultaneously, significantly simplifying the pipeline, allowing greater ease of practical use compared to NLNL. With a simple semi-supervised training technique, our method achieves state-of-the-art accuracy for noisy data classification based on the superior filtering ability. | - |
dc.language | English | - |
dc.publisher | IEEE Computer Society | - |
dc.title | Joint Negative and Positive Learning for Noisy Labels | - |
dc.type | Conference | - |
dc.identifier.wosid | 000742075007049 | - |
dc.identifier.scopusid | 2-s2.0-85123167151 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 9437 | - |
dc.citation.endingpage | 9446 | - |
dc.citation.publicationname | 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Virtual | - |
dc.identifier.doi | 10.1109/CVPR46437.2021.00932 | - |
dc.contributor.localauthor | Kim, Junmo | - |
dc.contributor.nonIdAuthor | Kim, Youngdong | - |
dc.contributor.nonIdAuthor | Yun, Juseung | - |
dc.contributor.nonIdAuthor | Shon, Hyounguk | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.