Any-Class Presence Likelihood for Robust Multi-Label Classification with Abundant Negative Data
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908395964989440 |
|---|---|
| author | Tissera, Dumindu Awadallah, Omar Danish, Muhammad Umair Sadhu, Ayan Grolinger, Katarina |
| author_facet | Tissera, Dumindu Awadallah, Omar Danish, Muhammad Umair Sadhu, Ayan Grolinger, Katarina |
| contents | Multi-label Classification (MLC) assigns an instance to one or more non-exclusive classes. A challenge arises when the dataset contains a large proportion of instances with no assigned class, referred to as negative data, which can overwhelm the learning process and hinder the accurate identification and classification of positive instances. Nevertheless, it is common in MLC applications such as industrial defect detection, agricultural disease identification, and healthcare diagnosis to encounter large amounts of negative data. Assigning a separate negative class to these instances further complicates the learning objective and introduces unnecessary redundancies. To address this challenge, we redesign standard MLC loss functions by deriving a likelihood of any class being present, formulated by a normalized weighted geometric mean of the predicted class probabilities. We introduce a regularization parameter that controls the relative contribution of the absent class probabilities to the any-class presence likelihood in positive instances. The any-class presence likelihood complements the multi-label learning by encouraging the network to become more aware of implicit positive instances and improve the label classification within those positive instances. Experiments on large-scale datasets with negative data: SewerML, modified COCO, and ChestX-ray14, across various networks and base loss functions show that our loss functions consistently improve MLC performance of their standard loss counterparts, achieving gains of up to 6.01 percentage points in F1, 8.06 in F2, and 3.11 in mean average precision, all without additional parameters or computational complexity. Code available at: https://github.com/ML-for-Sensor-Data-Western/gmean-mlc |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05721 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Any-Class Presence Likelihood for Robust Multi-Label Classification with Abundant Negative Data Tissera, Dumindu Awadallah, Omar Danish, Muhammad Umair Sadhu, Ayan Grolinger, Katarina Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition 68T05 (Primary) 62H30 (Secondary) I.2.6; I.5.4 Multi-label Classification (MLC) assigns an instance to one or more non-exclusive classes. A challenge arises when the dataset contains a large proportion of instances with no assigned class, referred to as negative data, which can overwhelm the learning process and hinder the accurate identification and classification of positive instances. Nevertheless, it is common in MLC applications such as industrial defect detection, agricultural disease identification, and healthcare diagnosis to encounter large amounts of negative data. Assigning a separate negative class to these instances further complicates the learning objective and introduces unnecessary redundancies. To address this challenge, we redesign standard MLC loss functions by deriving a likelihood of any class being present, formulated by a normalized weighted geometric mean of the predicted class probabilities. We introduce a regularization parameter that controls the relative contribution of the absent class probabilities to the any-class presence likelihood in positive instances. The any-class presence likelihood complements the multi-label learning by encouraging the network to become more aware of implicit positive instances and improve the label classification within those positive instances. Experiments on large-scale datasets with negative data: SewerML, modified COCO, and ChestX-ray14, across various networks and base loss functions show that our loss functions consistently improve MLC performance of their standard loss counterparts, achieving gains of up to 6.01 percentage points in F1, 8.06 in F2, and 3.11 in mean average precision, all without additional parameters or computational complexity. Code available at: https://github.com/ML-for-Sensor-Data-Western/gmean-mlc |
| title | Any-Class Presence Likelihood for Robust Multi-Label Classification with Abundant Negative Data |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition 68T05 (Primary) 62H30 (Secondary) I.2.6; I.5.4 |
| url | https://arxiv.org/abs/2506.05721 |