2020 Fiscal Year Research-status Report
Machine Learning on Large Graphs
| Project/Area Number |
18K11434
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2022-03-31
|
| Keywords | large graph / Convex clustering / graph neural networks |
|---|
| Outline of Annual Research Achievements |
This year, we are still working toward the main goal of learning sound models of graphs and their applications. We have found applications of graphs and sparse structured data in different situations. One is the case of sparse data in Bayesian streaming learning.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
We are still stuck at the main goal of learning sound models of graphs. However, recent advances in learning on graphs offered some hope. While it is difficult to prove the soundness of graph models, one can prove its value in extreme cases. We are planning on this direction, to prove the soundness of these models on semi-supervised learning with very few labelled training data and learning representations of nodes on graphs.
|
| Strategy for Future Research Activity |
We are still in search of the sound models of graphs as the main goal. Other than that, we will find applications of graph in different situations such as in convex clustering.
|
| Causes of Carryover |
Due to covid-19 pandemic, we could not use the research budget as planned.
|
