研究課題/領域番号 |
22KJ1537
|
補助金の研究課題番号 |
21J23611 (2021-2022)
|
研究種目 |
特別研究員奨励費
|
配分区分 | 基金 (2023) 補助金 (2021-2022) |
応募区分 | 国内 |
審査区分 |
小区分15010:素粒子、原子核、宇宙線および宇宙物理に関連する理論
|
研究機関 | 国立天文台 (2023) 名古屋大学 (2021-2022) |
研究代表者 |
COORAY Suchetha 国立天文台, 科学研究部, 特別研究員(PD)
|
研究期間 (年度) |
2023-03-08 – 2024-03-31
|
研究課題ステータス |
交付 (2023年度)
|
配分額 *注記 |
2,200千円 (直接経費: 2,200千円)
2023年度: 700千円 (直接経費: 700千円)
2022年度: 700千円 (直接経費: 700千円)
2021年度: 800千円 (直接経費: 800千円)
|
キーワード | Galaxies / Magnetism / Machine Learning |
研究開始時の研究の概要 |
Propose a novel approach to develop a comprehensive model for the multi-scale physics of galaxies. Utilize a combination of multi-wavelength emissions, spatial information, and magnetic field data, augmented by advanced unsupervised machine learning techniques, to extract valuable insights from observational data. The primary objective is to establish a unified model that incorporates magnetism information alongside the processes influencing star formation. This endeavor aims to bridge gaps in our understanding of galaxies, enabling a more holistic view of their formation and evolution.
|
研究実績の概要 |
Over the past year, my research has focused on three primary areas: the Galaxy Manifold and star formation history estimation. Utilizing unsupervised machine learning, I successfully identified two fundamental parameters that accurately represent galaxies, aligning with current understanding. Furthermore, I developed a method employing generative models to simulate star formation histories, enabling a better match with observations and advancing our comprehension of galaxy formation history. Additionally, I collaborated with the SKA-Japan Science working group on the cosmic magnetism project, exploring the observability of galactic magnetism through simulations and investigating alternative approaches to disentangle complex magnetic field structures.
|
現在までの達成度 (区分) |
現在までの達成度 (区分)
1: 当初の計画以上に進展している
理由
At present, the findings from my Galaxy Manifold research have been documented in a forthcoming paper, which consolidates the results of the multi-wavelength analysis of local galaxies. The paper highlights the consistent outcomes obtained using unsupervised machine learning techniques and their implications for our understanding of galaxy evolution within the vast Universe. Simultaneously, I am in the final stages of preparing a comprehensive publication on my innovative method for star formation history estimation. This approach, leveraging generative models, has shown promising results in generating simulation-like histories that closely match observations, overcoming the limitations of observational data alone.
|
今後の研究の推進方策 |
With the upcoming publication on the Galaxy Manifold research, I aim to disseminate the findings widely within the scientific community, fostering discussions and encouraging collaborations in the field of galactic analysis. Additionally, I have established a strong network of esteemed researchers in galaxy formation and machine learning. Moving forward, I plan to leverage this network and collaborate on developing a machine learning algorithm for galaxy formation that integrates multi-wavelength and multi-epoch observations. Furthermore, I will actively engage with researchers from renowned institutions such as Stanford University to explore alternative probes and innovative approaches to accurately unravel the complexities associated with galactic magnetism.
|