# Training image classifiers using Semi-Weak Label Data

arXiv: https://arxiv.org/abs/2103.10608 · HTML: https://mlsanity.com/paper/2103.10608/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 44 references that are in the map.
- First posted on arXiv 19 Mar 2021.

## Builds on (most cited first, 15 of 15)

- [Multiple Instance Learning: A Survey of Problem Characteristics and Applications](https://mlsanity.com/paper/1612.03365/) (2016)
- [Representation Learning by Learning to Count](https://mlsanity.com/paper/1708.06734/) (2017)
- [3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization](https://mlsanity.com/paper/1908.08216/) (2019)
- [Align2Ground: Weakly Supervised Phrase Grounding Guided by Image-Caption Alignment](https://mlsanity.com/paper/1903.11649/) (2019)
- [Learning from Label Proportions with Consistency Regularization](https://mlsanity.com/paper/1910.13188/) (2019)
- [Deep multi-class learning from label proportions](https://mlsanity.com/paper/1905.12909/) (2019)
- [Co-training for Demographic Classification Using Deep Learning from Label Proportions](https://mlsanity.com/paper/1709.04108/) (2017)
- [C-WSL: Count-guided Weakly Supervised Localization](https://mlsanity.com/paper/1711.05282/) (2017)
- [Deep Learning from Label Proportions for Emphysema Quantification](https://mlsanity.com/paper/1807.08601/) (2018)
- [Explanation-based Weakly-supervised Learning of Visual Relations with Graph Networks](https://mlsanity.com/paper/2006.09562/) (2020)
- [A Closer Look at Weak Label Learning for Audio Events](https://mlsanity.com/paper/1804.09288/) (2018)
- [Visual Recognition by Counting Instances: A Multi-Instance Cardinality Potential Kernel](https://mlsanity.com/paper/1502.02063/) (2015)
- [Learning from Multiple Corrupted Sources, with Application to Learning from Label Proportions](https://mlsanity.com/paper/1910.04665/) (2019)
- [On Classification with Bags, Groups and Sets](https://mlsanity.com/paper/1406.0281/) (2014)
- [Weakly Supervised Scalable Audio Content Analysis](https://mlsanity.com/paper/1606.03664/) (2016)

## Built on by (most cited first, 2 of 2)

- [Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations](https://mlsanity.com/paper/2305.12715/) (2023)
- [Importance of negative sampling in weak label learning](https://mlsanity.com/paper/2309.13227/) (2023)

## Abstract (arXiv)

In Multiple Instance learning (MIL), weak labels are provided at the bag level with only presence/absence information known. However, there is a considerable gap in performance in comparison to a fully supervised model, limiting the practical applicability of MIL approaches. Thus, this paper introduces a novel semi-weak label learning paradigm as a middle ground to mitigate the problem. We define semi-weak label data as data where we know the presence or absence of a given class and the exact count of each class as opposed to knowing the label proportions. We then propose a two-stage framework to address the problem of learning from semi-weak labels. It leverages the fact that counting information is non-negative and discrete. Experiments are conducted on generated samples from CIFAR-10. We compare our model with a fully-supervised setting baseline, a weakly-supervised setting baseline and learning from pro-portion (LLP) baseline. Our framework not only outperforms both baseline models for MIL-based weakly super-vised setting and learning from proportion setting, but also gives comparable results compared to the fully supervised model. Further, we conduct thorough ablation studies to analyze across datasets and variation with batch size, losses architectural changes, bag size and regularization