Rebalancing Multi-Label Class-Incremental Learning

• URL: http://arxiv.org/abs/2408.12161v1
• Date: Thu, 22 Aug 2024 07:04:22 GMT
• Title: Rebalancing Multi-Label Class-Incremental Learning
• Authors: Kaile Du, Yifan Zhou, Fan Lyu, Yuyang Li, Junzhou Xie, Yixi Shen, Fuyuan Hu, Guangcan Liu,
• Abstract summary: We propose a Rebalance framework for both the Loss and Label levels (RebLL) AKD is proposed to rebalance at the loss level by emphasizing the negative label learning in classification loss. OR is designed for label rebalance, which restores the original class distribution in memory by online relabeling the missing classes.
• Score: 17.921329790170265
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-label class-incremental learning (MLCIL) is essential for real-world multi-label applications, allowing models to learn new labels while retaining previously learned knowledge continuously. However, recent MLCIL approaches can only achieve suboptimal performance due to the oversight of the positive-negative imbalance problem, which manifests at both the label and loss levels because of the task-level partial label issue. The imbalance at the label level arises from the substantial absence of negative labels, while the imbalance at the loss level stems from the asymmetric contributions of the positive and negative loss parts to the optimization. To address the issue above, we propose a Rebalance framework for both the Loss and Label levels (RebLL), which integrates two key modules: asymmetric knowledge distillation (AKD) and online relabeling (OR). AKD is proposed to rebalance at the loss level by emphasizing the negative label learning in classification loss and down-weighting the contribution of overconfident predictions in distillation loss. OR is designed for label rebalance, which restores the original class distribution in memory by online relabeling the missing classes. Our comprehensive experiments on the PASCAL VOC and MS-COCO datasets demonstrate that this rebalancing strategy significantly improves performance, achieving new state-of-the-art results even with a vanilla CNN backbone.

Related papers

• Confidence Self-Calibration for Multi-Label Class-Incremental Learning [21.104984143597882]
  Partial label challenge in Multi-Label Class-Incremental Learning (MLCIL) arises when only the new classes are labeled during training. This issue leads to a proliferation of false-positive errors due to erroneously high confidence multi-label predictions. We propose a Confidence Self-Calibration (CSC) approach to refine multi-label confidence calibration in MLCIL.
  arXiv  Detail & Related papers  (2024-03-19T09:14:52Z)
• Gradient Reweighting: Towards Imbalanced Class-Incremental Learning [8.438092346233054]
  Class-Incremental Learning (CIL) trains a model to continually recognize new classes from non-stationary data. A major challenge of CIL arises when applying to real-world data characterized by non-uniform distribution. We show that this dual imbalance issue causes skewed gradient updates with biased weights in FC layers, thus inducing over/under-fitting and catastrophic forgetting in CIL.
  arXiv  Detail & Related papers  (2024-02-28T18:08:03Z)
• PercentMatch: Percentile-based Dynamic Thresholding for Multi-Label Semi-Supervised Classification [64.39761523935613]
  We propose a percentile-based threshold adjusting scheme to dynamically alter the score thresholds of positive and negative pseudo-labels for each class during the training. We achieve strong performance on Pascal VOC2007 and MS-COCO datasets when compared to recent SSL methods.
  arXiv  Detail & Related papers  (2022-08-30T01:27:48Z)
• Transductive CLIP with Class-Conditional Contrastive Learning [68.51078382124331]
  We propose Transductive CLIP, a novel framework for learning a classification network with noisy labels from scratch. A class-conditional contrastive learning mechanism is proposed to mitigate the reliance on pseudo labels. ensemble labels is adopted as a pseudo label updating strategy to stabilize the training of deep neural networks with noisy labels.
  arXiv  Detail & Related papers  (2022-06-13T14:04:57Z)
• Phased Progressive Learning with Coupling-Regulation-Imbalance Loss for Imbalanced Classification [11.673344551762822]
  Deep neural networks generally perform poorly with datasets that suffer from quantity imbalance and classification difficulty imbalance between different classes. A phased progressive learning schedule was proposed for smoothly transferring the training emphasis from representation learning to upper classifier training. Our code will be open source soon.
  arXiv  Detail & Related papers  (2022-05-24T14:46:39Z)
• Distribution-Aware Semantics-Oriented Pseudo-label for Imbalanced Semi-Supervised Learning [80.05441565830726]
  This paper addresses imbalanced semi-supervised learning, where heavily biased pseudo-labels can harm the model performance. We propose a general pseudo-labeling framework to address the bias motivated by this observation. We term the novel pseudo-labeling framework for imbalanced SSL as Distribution-Aware Semantics-Oriented (DASO) Pseudo-label.
  arXiv  Detail & Related papers  (2021-06-10T11:58:25Z)
• PLM: Partial Label Masking for Imbalanced Multi-label Classification [59.68444804243782]
  Neural networks trained on real-world datasets with long-tailed label distributions are biased towards frequent classes and perform poorly on infrequent classes. We propose a method, Partial Label Masking (PLM), which utilizes this ratio during training. Our method achieves strong performance when compared to existing methods on both multi-label (MultiMNIST and MSCOCO) and single-label (imbalanced CIFAR-10 and CIFAR-100) image classification datasets.
  arXiv  Detail & Related papers  (2021-05-22T18:07:56Z)
• Disentangling Sampling and Labeling Bias for Learning in Large-Output Spaces [64.23172847182109]
  We show that different negative sampling schemes implicitly trade-off performance on dominant versus rare labels. We provide a unified means to explicitly tackle both sampling bias, arising from working with a subset of all labels, and labeling bias, which is inherent to the data due to label imbalance.
  arXiv  Detail & Related papers  (2021-05-12T15:40:13Z)
• Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets [98.74153364118898]
  We present a new loss function called Distribution-Balanced Loss for the multi-label recognition problems that exhibit long-tailed class distributions. The Distribution-Balanced Loss tackles these issues through two key modifications to the standard binary cross-entropy loss. Experiments on both Pascal VOC and COCO show that the models trained with this new loss function achieve significant performance gains.
  arXiv  Detail & Related papers  (2020-07-19T11:50:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.