PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909404738093056 |
|---|---|
| author | Liu, Xiaoyu Zhou, Beitong Yue, Zuogong Cheng, Cheng |
| author_facet | Liu, Xiaoyu Zhou, Beitong Yue, Zuogong Cheng, Cheng |
| contents | Recently, the usage of Contrastive Representation Learning (CRL) as a pre-training technique improves the performance of learning with noisy labels (LNL) methods. However, instead of pre-training, when trivially combining CRL loss with LNL methods as an end-to-end framework, the empirical experiments show severe degeneration of the performance. We verify through experiments that this issue is caused by optimization conflicts of losses and propose an end-to-end \textbf{PLReMix} framework by introducing a Pseudo-Label Relaxed (PLR) contrastive loss. This PLR loss constructs a reliable negative set of each sample by filtering out its inappropriate negative pairs, alleviating the loss conflicts by trivially combining these losses. The proposed PLR loss is pluggable and we have integrated it into other LNL methods, observing their improved performance. Furthermore, a two-dimensional Gaussian Mixture Model is adopted to distinguish clean and noisy samples by leveraging semantic information and model outputs simultaneously. Experiments on multiple benchmark datasets demonstrate the effectiveness of the proposed method. Code is available at \url{https://github.com/lxysl/PLReMix}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_17589 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning Liu, Xiaoyu Zhou, Beitong Yue, Zuogong Cheng, Cheng Computer Vision and Pattern Recognition Recently, the usage of Contrastive Representation Learning (CRL) as a pre-training technique improves the performance of learning with noisy labels (LNL) methods. However, instead of pre-training, when trivially combining CRL loss with LNL methods as an end-to-end framework, the empirical experiments show severe degeneration of the performance. We verify through experiments that this issue is caused by optimization conflicts of losses and propose an end-to-end \textbf{PLReMix} framework by introducing a Pseudo-Label Relaxed (PLR) contrastive loss. This PLR loss constructs a reliable negative set of each sample by filtering out its inappropriate negative pairs, alleviating the loss conflicts by trivially combining these losses. The proposed PLR loss is pluggable and we have integrated it into other LNL methods, observing their improved performance. Furthermore, a two-dimensional Gaussian Mixture Model is adopted to distinguish clean and noisy samples by leveraging semantic information and model outputs simultaneously. Experiments on multiple benchmark datasets demonstrate the effectiveness of the proposed method. Code is available at \url{https://github.com/lxysl/PLReMix}. |
| title | PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.17589 |