Key Instance Detection in Multi-Instance Learning


Guoqing Liu, Jianxin Wu, Zhi-Hua Zhou ;
Proceedings of the Asian Conference on Machine Learning, PMLR 25:253-268, 2012.


The goal of traditional multi-instance learning (MIL) is to predict the labels of the bags, whereas in many real applications, it is desirable to get the instance labels, especially the labels of key instances that trigger the bag labels, in addition to getting bag labels. Such a problem has been largely unexplored before. In this paper, we formulate the Key Instance Detection (KID) problem, and propose a voting framework (VF) solution to KID. The key of VF is to exploit the relationship among instances, represented by a citer kNN graph. This graph is different from commonly used nearest neighbor graphs, but is suitable for KID. Experiments validate the effectiveness of VF for KID. Additionally, VF also outperforms state-of-the-art MIL approaches on the performance of bag label prediction.

Related Material