Reinforced GNNs for Multiple Instance Learning
- 30 April 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (4) , 6693-6707
Abstract
Multiple instance learning (MIL) trains models from bags of instances, where each bag contains multiple instances, and only bag-level labels are available for supervision. The application of graph neural networks (GNNs) in capturing intrabag topology effectively improves MIL. Existing GNNs usually require filtering low-confidence edges among instances and adapting graph neural architectures to new bag structures. However, such asynchronous adjustments to structure and architecture are tedious and ignore their correlations. To tackle these issues, we propose a reinforced GNN framework for MIL (RGMIL), pioneering the exploitation of multiagent deep reinforcement learning (MADRL) in MIL tasks. MADRL enables the flexible definition or extension of factors that influence bag graphs or GNNs and provides synchronous control over them. Moreover, MADRL explores structure-to-architecture correlations while automating adjustments. Experimental results on multiple MIL datasets demonstrate that RGMIL achieves the best performance with excellent explainability. The code and data are available at https://github.com/RingBDStack/RGMIL.Keywords
Funding Information
- National Key Research and Development Program of China (2021YFB1714800)
- NSFC (2021YFB1714800, 62322202, 61932002, U21B2027, U23A20388, 62266028)
- Beijing Natural Science Foundation (4222030)
- Basic and Applied Basic Research Foundation of Guangdong Province (2023B1515120020)
- Shijiazhuang Science and Technology Plan (231130459A)
This publication has 44 references indexed in Scilit:
- Deep reinforcement learning guided graph neural networks for brain network analysisNeural Networks, 2022
- Node-aligned Graph Convolutional Network for Whole-slide Image Representation and Classification2022
- Multi-instance learning of graph neural networks for aqueous pKa predictionBioinformatics, 2021
- DMIL-IsoFun: predicting isoform function using deep multi-instance learningBioinformatics, 2021
- Predicting Lymph Node Metastasis Using Histopathological Images Based on Multiple Instance Learning With Deep Graph Convolution2020
- Representation Learning of Histopathology Images using Graph Neural Networks2020
- Revisiting multiple instance neural networksPattern Recognition, 2018
- Computer-aided diagnosis from weak supervision: A benchmarking studyComputerized Medical Imaging and Graphics, 2015
- Human-level control through deep reinforcement learningNature, 2015
- Identification and expansion of human colon-cancer-initiating cellsNature, 2006