Label-efficient Machine Learning

1. Learning from Open-set Noisy Labels based on Multi-prototype Modeling

In this paper, we propose a novel method to address the challenge of learning deep neural network models in the presence of open-set noisy labels, which include mislabeled samples from out-of-distribution categories. Previous methods relied on the …

2. Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation

3. Weakly Supervised Semantic Segmentation via Alternate Self-Dual Teaching

4. RankMatch: Fostering Confidence and Consistency in Learning with Noisy Labels

5. Reliable Mutual Distillation for Medical Image Segmentation under Imperfect Annotations

6. Gradient-Rebalanced Uncertainty Minimization for Cross-Site Adaptation of Medical Image Segmentation

7. Double-Check Soft Teacher for Semi-Supervised Object Detection

In the semi-supervised object detection task, due to the scarcity of labeled data and the diversity and complexity of objects to be detected, the quality of pseudo-labels generated by existing methods for unlabeled data is relatively low, which …

8. Cross-level Contrastive Learning and Consistency Constraint for Semi-supervised Medical Image Segmentation

9. Interactive Dual-Model Learning for Semi-supervised Medical Image Segmentation (in Chinese)

10. Weakly Supervised Disease Localization in Chest X-rays via Looking into Image Relations