# Pushing One Pair of Labels Apart Each Time in Multi-Label Learning: From Single Positive to Full Labels

arXiv: https://arxiv.org/abs/2302.14695 · HTML: https://mlsanity.com/paper/2302.14695/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 67 references that are in the map.
- First posted on arXiv 28 Feb 2023.

## Builds on (most cited first, 30 of 32)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Early-Learning Regularization Prevents Memorization of Noisy Labels](https://mlsanity.com/paper/2007.00151/) (2020)
- [Asymmetric Loss For Multi-Label Classification](https://mlsanity.com/paper/2009.14119/) (2020)
- [Multi-Label Image Recognition with Graph Convolutional Networks](https://mlsanity.com/paper/1904.03582/) (2019)
- [Normalized Loss Functions for Deep Learning with Noisy Labels](https://mlsanity.com/paper/2006.13554/) (2020)
- [Does label smoothing mitigate label noise?](https://mlsanity.com/paper/2003.02819/) (2020)
- [CNN-RNN: A Unified Framework for Multi-label Image Classification](https://mlsanity.com/paper/1604.04573/) (2016)
- [General Multi-label Image Classification with Transformers](https://mlsanity.com/paper/2011.14027/) (2020)
- [Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets](https://mlsanity.com/paper/2007.09654/) (2020)
- [Meta Label Correction for Noisy Label Learning](https://mlsanity.com/paper/1911.03809/) (2019)
- [The Emerging Trends of Multi-Label Learning](https://mlsanity.com/paper/2011.11197/) (2020)
- [Delving Deep into Label Smoothing](https://mlsanity.com/paper/2011.12562/) (2020)
- [Learning Spatial Regularization with Image-level Supervisions for Multi-label Image Classification](https://mlsanity.com/paper/1702.05891/) (2017)
- [FINE Samples for Learning with Noisy Labels](https://mlsanity.com/paper/2102.11628/) (2021)
- [Multi-Label Learning from Single Positive Labels](https://mlsanity.com/paper/2106.09708/) (2021)
- [Fast Multi-Resolution Transformer Fine-tuning for Extreme Multi-label Text Classification](https://mlsanity.com/paper/2110.00685/) (2021)
- [Learning with Noisy Labels via Sparse Regularization](https://mlsanity.com/paper/2108.00192/) (2021)
- [Large Loss Matters in Weakly Supervised Multi-Label Classification](https://mlsanity.com/paper/2206.03740/) (2022)
- [Multi-Label Image Classification via Knowledge Distillation from Weakly-Supervised Detection](https://mlsanity.com/paper/1809.05884/) (2018)
- [A Unified View of Multi-Label Performance Measures](https://mlsanity.com/paper/1609.00288/) (2016)
- [Hyperbolic Interaction Model For Hierarchical Multi-Label Classification](https://mlsanity.com/paper/1905.10802/) (2019)
- [Acknowledging the Unknown for Multi-label Learning with Single Positive Labels](https://mlsanity.com/paper/2203.16219/) (2022)
- [One Positive Label is Sufficient: Single-Positive Multi-Label Learning with Label Enhancement](https://mlsanity.com/paper/2206.00517/) (2022)
- [A Concise yet Effective model for Non-Aligned Incomplete Multi-view and Missing Multi-label Learning](https://mlsanity.com/paper/2005.00976/) (2020)
- [ZLPR: A Novel Loss for Multi-label Classification](https://mlsanity.com/paper/2208.02955/) (2022)
- [Multi-Label Activity Recognition using Activity-specific Features and Activity Correlations](https://mlsanity.com/paper/2009.07420/) (2020)
- [Evaluating Multi-label Classifiers with Noisy Labels](https://mlsanity.com/paper/2102.08427/) (2021)

## Abstract (arXiv)

In Multi-Label Learning (MLL), it is extremely challenging to accurately annotate every appearing object due to expensive costs and limited knowledge. When facing such a challenge, a more practical and cheaper alternative should be Single Positive Multi-Label Learning (SPMLL), where only one positive label needs to be provided per sample. Existing SPMLL methods usually assume unknown labels as negatives, which inevitably introduces false negatives as noisy labels. More seriously, Binary Cross Entropy (BCE) loss is often used for training, which is notoriously not robust to noisy labels. To mitigate this issue, we customize an objective function for SPMLL by pushing only one pair of labels apart each time to prevent the domination of negative labels, which is the main culprit of fitting noisy labels in SPMLL. To further combat such noisy labels, we explore the high-rankness of label matrix, which can also push apart different labels. By directly extending from SPMLL to MLL with full labels, a unified loss applicable to both settings is derived. Experiments on real datasets demonstrate that the proposed loss not only performs more robustly to noisy labels for SPMLL but also works well for full labels. Besides, we empirically discover that high-rankness can mitigate the dramatic performance drop in SPMLL. Most surprisingly, even without any regularization or fine-tuned label correction, only adopting our loss defeats state-of-the-art SPMLL methods on CUB, a dataset that severely lacks labels.