paper-with-me

홈 › Papers

Relation Modeling and Distillation for Learning with Noisy Labels

2024-05-30 · Xiaming Che, Junlin Zhang, Zhuang Qi, Xin Qi

Learning with noisy labels has become an effective strategy for enhancing the robustness of models, which enables models to better tolerate inaccurate data. Existing methods either focus on optimizing the loss function to mitigate the interference from noise, or design procedures to detect potential noise and correct errors. However, their effectiveness is often compromised in representation learning due to the dilemma where models overfit to noisy labels. To address this issue, this paper proposes a relation modeling and distillation framework that models inter-sample relationships via self-supervised learning and employs knowledge distillation to enhance understanding of latent associations, which mitigate the impact of noisy labels. Specifically, the proposed method, termed RMDNet, includes two main modules, where the relation modeling (RM) module implements the contrastive learning technique to learn representations of all data, an unsupervised approach that effectively eliminates the interference of noisy tags on feature extraction. The relation-guided representation learning (RGRL) module utilizes inter-sample relation learned from the RM module to calibrate the representation distribution for noisy samples, which is capable of improving the generalization of the model in the inference phase. Notably, the proposed RMDNet is a plug-and-play framework that can integrate multiple methods to its advantage. Extensive experiments were conducted on two datasets, including performance comparison, ablation study, in-depth analysis and case study. The results show that RMDNet can learn discriminative representations for noisy data, which results in superior performance than the existing methods.

📄 PDF Abstract BibTeX arXiv:2405.19606

Code (0)

등록된 구현이 없습니다.

Tasks

Contrastive LearningKnowledge DistillationLearning with noisy labelsRelationRepresentation LearningSelf-Supervised Learning

Methods 이 논문이 사용한 방법론

Focus 설명 없음
Contrastive Learning 설명 없음
Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Learning from Noisy Labels with Distillation

2017-03-07 · ICCV 2017 10 · Yuncheng Li, Jianchao Yang, Yale Song, Liangliang Cao 외

The ability of learning from noisy labels is very useful in many visual recognition tasks, as a vast amount of data with noisy labels are relatively easy to obtain. Traditionally, the label noises have been treated as st…

Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels

2022-10-17 · Brandon Smart, Gustavo Carneiro

Many state-of-the-art noisy-label learning methods rely on learning mechanisms that estimate the samples' clean labels during training and discard their original noisy labels. However, this approach prevents the learning…

Image ClassificationLearning with noisy labels

Federated Learning with Extremely Noisy Clients via Negative Distillation

2023-12-20 · Yang Lu, Lin Chen, Yonggang Zhang, Yiliang Zhang 외

Federated learning (FL) has shown remarkable success in cooperatively training deep models, while typically struggling with noisy labels. Advanced works propose to tackle label noise by a re-weighting strategy with a str…

Federated LearningKnowledge Distillation

FlyKD: Graph Knowledge Distillation on the Fly with Curriculum Learning

2024-03-16 · Eugene Ku

Knowledge Distillation (KD) aims to transfer a more capable teacher model's knowledge to a lighter student model in order to improve the efficiency of the model, making it faster and more deployable. However, the student…

Knowledge Distillation

Blind Knowledge Distillation for Robust Image Classification

2022-11-21 · Timo Kaiser, Lukas Ehmann, Christoph Reinders, Bodo Rosenhahn

Optimizing neural networks with noisy labels is a challenging task, especially if the label set contains real-world noise. Networks tend to generalize to reasonable patterns in the early training stages and overfit to sp…

Classificationimage-classificationImage ClassificationKnowledge Distillation+1