利用图神经网络进行多实例学习

Jun, 2019

Multiple instance learning with graph neural networks

Ming Tu, Jing Huang, Xiaodong He, Bowen Zhou

TL;DR本文提出一种基于图神经网络的多实例学习算法，将每个数据包视为图，并使用图神经网络学习包嵌入，利用实例之间的结构信息来预测标签。实验证明，该算法在多个常用数据集上达到了最先进的效果，且不失模型的可解释性。

Abstract

multiple instance learning (MIL) aims to learn the mapping between a bag of instances and the bag-level label. In this paper, we propose a new end-to-end graph neural network (GNN) based algorithm for MIL: we tre