Neural Network based Graph Learning: Model Evolution and Real-World Application
Project/Area Number |
21K12042
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 61030:Intelligent informatics-related
|
Research Institution | National Institute of Advanced Industrial Science and Technology |
Principal Investigator |
劉 欣 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 主任研究員 (20803935)
|
Project Period (FY) |
2021-04-01 – 2024-03-31
|
Project Status |
Granted (Fiscal Year 2022)
|
Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2023: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2022: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2021: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
Keywords | Graph Neural Network / Graph Embedding / Node Classification / Dynamic Graph / Imbalanced Graph / Multi-Label Graph / Geometric Learning / Graph Mining / Link Prediction / graph neural network / graph embedding / graph analysis / social network / complex network |
Outline of Research at the Start |
Graph learning has been advancing rapidly. However, several intrinsic defects hinder this technology from achieving immense success. We aim to remedy these defects and promote model evolutions. The outcome will improve the practicability of this technology and facilitate its extensive usage.
|
Outline of Annual Research Achievements |
We have achieved four main outcomes. a) We introduced several key design strategies for GNNs and proposed new models. b) We proposed single-level aggregation scheme for heterogeneous graph embedding, which addresses the down-weighting issue associated with the current bi-level scheme. c) We proposed data generation strategy and new GNN models for tackling the imbalanced graph node classification problem. d) We employed graph learning approach for predicting the donations on live streaming platforms.
|
Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
The current progress of the project is going well. There are no delays. The finished work has been published in prestigious conferences and journals. We are working on learning graphs in noisy environments.
|
Strategy for Future Research Activity |
We will focus on the following problems: 1) graph learning in noisy environments, 2) GNN for dynamic graphs and its applications.
|
Report
(2 results)
Research Products
(32 results)