研究課題/領域番号 |
21K17808
|
研究機関 | 九州大学 |
研究代表者 |
|
研究期間 (年度) |
2021-04-01 – 2023-03-31
|
キーワード | Time series / Temporal neural network / Dynamic programming |
研究実績の概要 |
In 2021, a number of publications at international conferences and journals were achieved. As part of the goals of the research, I sought to develop dynamic neural networks, networks that could tackle real world data, and research that tackled fundamental issues in deep learning. I was successfully able to publish papers and develop collaborations in these topics. For example, a paper in PLOS One provides a deep survey on data augmentation for time series classification, a paper published at ICASSP deals with varying length time series, and a paper under review at ICPR works with dynamic neural networks. In addition, I have collaborated with interdisciplinary fields such as remote sensing and natural language processing.
There were three international journals and 11 conference presentations.
|
現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
The research had a high level of success, especially when pertaining to the algorithms developed and the publications.
|
今後の研究の推進方策 |
This research and the collaborations will continue beyond the period of the grant. Currently, I am writing a journal paper based on the results from the research performed under this grand. Furthermore, multiple students under my supervision have papers under review or in the development stage.
In the future, this research topic will be extended. Not only is the ideas of the research important, but the application of it can be used for many other topics, such as multi-modal models.
|
次年度使用額が生じた理由 |
The main equipment needed for the next steps in the research is still in the process of being purchases. A request for the equipment was submitted, but the availability of the equipment is limited and is waiting. Also, due to the COVID situation, there were less travel expenses compared to previous years. This coming year will have more travel expenses.
|