研究生: |
韓惟晞 Han, Wei-Hsi |
---|---|
論文名稱: |
卷積神經網路與孿生神經網路在系外行星研究的應用 The Applications of Convolutional Neural Network and Siamese Neural Network on the Exoplanet Researches |
指導教授: |
葉麗琴
Yeh, Li-Chin |
口試委員: |
江瑛貴
Jiang, Ing-Guey 陳賢修 Chen, Shyan-Shiou |
學位類別: |
碩士 Master |
系所名稱: |
理學院 - 計算與建模科學研究所 Institute of Computational and Modeling Science |
論文出版年: | 2024 |
畢業學年度: | 112 |
語文別: | 中文 |
論文頁數: | 58 |
中文關鍵詞: | 凌星法 、系外行星 、機器學習 、卷積神經網路 、孿生神經網路 |
外文關鍵詞: | Transit Method, Exoplanet, Machine Learning, Convolutional Neural Network, Siamese Neural Network |
相關次數: | 點閱:89 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究旨在使用機器學習方法在TESS資料庫中尋找候選系外行星。我們首先將所有原始光曲線數據進行資料處理並作為雜訊,將資料分為兩個資料組別,接著加入 (Mandel & Agol, 2002) [4] 的凌星模組來模擬凌星現象,透過建構好的三種神經網路模型,分別為卷積神經網路 (Convolutional Neural Network, CNN)、孿生神經網路–對比損失 (Siamese Neural Network - Contrastive Loss) 與孿生神經網路–三元組損失 (Siamese Neural Network - Triplet Loss) 進行模型訓練和預測,並且分別選取最佳的模型作為我們尋找候選系外行星之模型。最後,我們在原始光曲線中尋找凌星週期在1至2天的候選系外行星,並透過模型預測找到了多個可能的候選系外行星。其中,訓練資料第一組別總共找到7個候選系外行星;而在訓練資料第二組別中,總共找到了4個候選系外行星。
We use machine learning methods to search for candidate exoplanets in the TESS database. We first processed all raw light curve data to reduce noise and divided the data into two categories. We then incorporated the transit modeling module from (Mandel & Agol, 2002) [4] to simulate transit phenomena. Using three constructed neural network models which are Convolutional Neural Network (CNN), Siamese Neural Network with Contrastive Loss, and Siamese Neural Network with Triplet Loss, we conducted models training and prediction. Finally, we selected the best-performing model from these experiments for exoplanet candidate search. Subsequently, we identified candidate exoplanets with transit periods between 1 to 2 days in the original light curves. Through model predictions, multiple potential candidate exoplanets were identified, with a total of seven candidates found in the first category and four candidates in the second category.
[1] G, Prithivraj & Kumari, Alka. (2023). Identification and Classification of Exoplanets Using Machine Learning Techniques. eprint arXiv:2305.09596.
[2] Hadsell, R., Chopra, S., & LeCun, Y. (2006). Dimensionality Reduction by Learning an Invariant Mapping. 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.
[3] Keras - Image similarity estimation using a Siamese Network with a contrastive loss: https://keras.io/examples/vision/siamese_contrastive/.
[4] Mandel, K., & Agol, E. (2002). Analytic Light Curves for Planetary Transit Searches. The Astrophysical Journal, 580(2): L171.
[5] Schroff, F., Kalenichenko, D., & Philbin, J. (2015). FaceNet: A Unified Embedding for Face Recognition and Clustering. 2015 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.
[6] TESS - FFI/TP/LC Bulk Downloads: https://archive.stsci.edu/tess/bulk_downloads/bulk_downloads_ffi-tp-lc-dv.html.
[7] Yeh, L. -C., & Jiang, I. –G. (2020). Searching for Exoplanet Transits from BRITE Data through a Machine Learning Technique. Publications of the Astronomical Society of the Pacific, 133(1019): 014401.