Tackling real-world time series using dynamic neural networks
Project/Area Number |
23K16949
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 61030:Intelligent informatics-related
|
Research Institution | Kyushu University |
Principal Investigator |
|
Project Period (FY) |
2023-04-01 – 2025-03-31
|
Project Status |
Granted (Fiscal Year 2023)
|
Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2024: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2023: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
|
Keywords | Time series / Temporal neural network / Dynamic programming |
Outline of Research at the Start |
Dynamic neural networks are artificial neural networks that can adapt their structures based on the input during inference. This research will explore dynamic feature alignment in order to dynamically warp the structure and parameters of temporal neural networks.
|
Outline of Annual Research Achievements |
This research resulted in many international publications. There were three international journal and five international conference presentations. Of the publications, two were in one of the top international journals in the proposals field, Pattern Recognition. The international conference papers were peer-reviewed and were part of many top conferences, such as ICCV and ICDAR. In addition, I have collaborated with interdisciplinary fields such as remote sensing, bioinformatics, and natural language processing.
|
Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
This research had success in many areas including feature representation, generation, and adversarial attacks.
|
Strategy for Future Research Activity |
The research will continue as planned. In addition, new collaborations and additional research topics will be developed. There are multiple papers under review by myself and my students.
|
Report
(1 results)
Research Products
(11 results)