Saved in:
Bibliographic Details
Main Authors: Zankov, Dmitry, Polishchuk, Pavlo, Sobieraj, Michal, Barbatti, Mario
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