Compressing Features for Learning With Noisy Labels

Chen, Yingyi; Hu, Shell Xu; Shen, Xi Sisi; Ai, Chunrong; Suykens, Johan A. K.

doi:10.1109/tnnls.2022.3186930

Cited by 13 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, for sample reweighting, Co-teaching [47] is a classical method that cross-updates its two base networks on the small-loss samples selected by its peer. Based on this, Nested Co-teaching [50,51]…”

Section: Related Workmentioning

confidence: 94%

Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer

Chen¹,

Shen²,

Liu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The success of Vision Transformer (ViT) in various computer vision tasks has promoted the everincreasing prevalence of this convolution-free network. The fact that ViT works on image patches makes it potentially relevant to the problem of jigsaw puzzle solving, which is a classical selfsupervised task aiming at reordering shuffled sequential image patches back to their natural form. Despite its simplicity, solving jigsaw puzzle has been demonstrated to be helpful for diverse tasks using Convolutional Neural Networks (CNNs), such as self-supervised feature representation learning, domain generalization, and fine-grained classification. In this paper, we explore solving jigsaw puzzle as a self-supervised auxiliary loss in ViT for image classification, named Jigsaw-ViT. We show two modifications that can make Jigsaw-ViT superior to standard ViT: discarding positional embeddings and masking patches randomly. Yet simple, we find that Jigsaw-ViT is able to improve both in generalization and robustness over the standard ViT, which is usually rather a trade-off. Experimentally, we show that adding the jigsaw puzzle branch provides better generalization than ViT on large-scale image classification on ImageNet. Moreover, the auxiliary task also improves robustness to noisy labels on Animal-10N, Food-101N, and Clothing1M as well as adversarial examples. Our implementation is available at https:// yingyichen-cyy.github.io/Jigsaw-ViT.

show abstract

Section: Related Workmentioning

confidence: 94%

Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer

Chen¹,

Shen²,

Liu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In particular, for sample reweighting, Co-teaching [31] is a classical method that cross-updates its two base networks on the small-loss samples selected by its peer. Based on this, Nested Co-teaching [32] improves performance by including compression regularization during the training.…”

Section: Related Workmentioning

confidence: 99%

“…The dataset is categorized into 14 classes, containing 1 , 0 0 0 , 0 0 0 training images with noise ratio ∼38% and 10 , 526 test images. We follow the preprocessing in [29,32] for this dataset.…”

Section: Robustness To Label Noisementioning

confidence: 99%

See 1 more Smart Citation

Jigsaw-ViT: Learning jigsaw puzzles in vision transformer

Chen

Shen

Liu

et al. 2023

Pattern Recognition Letters

View full text Add to dashboard Cite

“…(Song et al 2019, Zhang et al 2021, Chen et al 2022, Gao et al 2022, Xia et al 2023. We conducted our method on different backbones, including ResNet-18, ResNet-34 and VGG-19.…”

mentioning

confidence: 99%

Suppressing label noise in medical image classification using mixup attention and self-supervised learning

Gao,

Jiang,

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Deep neural networks (DNNs) have been widely applied in medical image classification and achieve remarkable classification performance. These achievements heavily depend on large-scale accurately annotated training data. However, label noise is inevitably introduced in the medical image annotation, as the labeling process heavily relies on the expertise and experience of annotators. Meanwhile, DNNs suffer from overfitting noisy labels, degrading the performance of models. Therefore, in this work, we innovatively devise noise-robust training approach to mitigate the adverse effects of noisy labels in medical image classification. Specifically, we incorporate contrastive learning and intra-group attention mixup strategies into the vanilla supervised learning. The contrastive learning for feature extractor helps to enhance visual representation of DNNs. The intra-group attention mixup module constructs groups and assigns self-attention weights for group-wise samples, and subsequently interpolates massive noisy-suppressed samples through weighted mixup operation. We conduct comparative experiments on both synthetic and real-world noisy medical datasets under various noise levels. Rigorous experiments validate that our noise-robust method with contrastive learning and attention mixup can effectively handle with label noise, and is superior to state-of-the-art methods. An ablation study also shows that both components contribute to boost model performance. The proposed method demonstrates its capability of curb label noise and has certain potential toward real-world clinic applications.

show abstract

Compressing Features for Learning With Noisy Labels

Cited by 13 publications

References 39 publications

Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer

Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer

Jigsaw-ViT: Learning jigsaw puzzles in vision transformer

Suppressing label noise in medical image classification using mixup attention and self-supervised learning

Contact Info

Product

Resources

About