Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.01287 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912740458627072 |
|---|---|
| author | Zankov, Dmitry Polishchuk, Pavlo Sobieraj, Michal Barbatti, Mario |
| author_facet | Zankov, Dmitry Polishchuk, Pavlo Sobieraj, Michal Barbatti, Mario |
| contents | We introduce milearn, a Python package for multi-instance learning (MIL) that follows the familiar scikit-learn fit/predict interface while providing a unified framework for both classical and neural-network-based MIL algorithms for regression and classification. The package also includes built-in hyperparameter optimization designed specifically for small MIL datasets, enabling robust model selection in data-scarce scenarios. We demonstrate the versatility of milearn across a broad range of synthetic MIL benchmark datasets, including digit classification and regression, molecular property prediction, and protein-protein interaction (PPI) prediction. Special emphasis is placed on the key instance detection (KID) problem, for which the package provides dedicated support. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01287 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | milearn: A Python Package for Multi-Instance Machine Learning Zankov, Dmitry Polishchuk, Pavlo Sobieraj, Michal Barbatti, Mario Machine Learning We introduce milearn, a Python package for multi-instance learning (MIL) that follows the familiar scikit-learn fit/predict interface while providing a unified framework for both classical and neural-network-based MIL algorithms for regression and classification. The package also includes built-in hyperparameter optimization designed specifically for small MIL datasets, enabling robust model selection in data-scarce scenarios. We demonstrate the versatility of milearn across a broad range of synthetic MIL benchmark datasets, including digit classification and regression, molecular property prediction, and protein-protein interaction (PPI) prediction. Special emphasis is placed on the key instance detection (KID) problem, for which the package provides dedicated support. |
| title | milearn: A Python Package for Multi-Instance Machine Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2512.01287 |