Deep Multi-Instance Learning with Dynamic Pooling
Abstract
End-to-end optimization of multi-instance learning (MIL) using neural networks is an important problem with many applications, in which a core issue is how to design a permutation-invariant pooling function without losing much instance-level information. Inspired by the dynamic routing in recent capsule networks, we propose a novel dynamic pooling function for MIL. It is an adaptive scheme for both key instance selection and modeling the contextual information among instances in a bag. The dynamic pooling iteratively updates the instance contribution to its bag. It is permutation-invariant and can interpret instance-to-bag relationship. The proposed dynamic pooling based multi-instance neural network has been validated on many MIL tasks and outperforms other MIL methods.
Cite
Text
Yan et al. "Deep Multi-Instance Learning with Dynamic Pooling." Proceedings of The 10th Asian Conference on Machine Learning, 2018.Markdown
[Yan et al. "Deep Multi-Instance Learning with Dynamic Pooling." Proceedings of The 10th Asian Conference on Machine Learning, 2018.](https://mlanthology.org/acml/2018/yan2018acml-deep/)BibTeX
@inproceedings{yan2018acml-deep,
title = {{Deep Multi-Instance Learning with Dynamic Pooling}},
author = {Yan, Yongluan and Wang, Xinggang and Guo, Xiaojie and Fang, Jiemin and Liu, Wenyu and Huang, Junzhou},
booktitle = {Proceedings of The 10th Asian Conference on Machine Learning},
year = {2018},
pages = {662-677},
volume = {95},
url = {https://mlanthology.org/acml/2018/yan2018acml-deep/}
}