Scholar
Miguel Martinho
Google Scholar ID: qG-OHOAAAAAJ
NASA Ames Research Center
Citations & Impact
Citations
167
H-index
5
i10-index
5
Publications
20
Co-authors
0
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- - Developed the ExoMiner pipeline and integrated it into an end-to-end process for the TESS mission, which queries SPOC TCEs, processes light curve and image data from MAST, and generates predictions for detected TCEs.
- - Adapted ExoMiner to TESS mission data, generating catalogs of Threshold Crossing Events (TCEs) and Community TESS Objects of Interest (CTOIs) from hundreds of thousands of transit signals.
- - As a main developer of ExoMiner, this method uses multi-modal data processing branches to mimic diagnostic tests by SMEs to reject false positive transit signals, and conducted additional statistical validation of 69 Kepler planet candidates.
Research Experience
- Serves as an applied machine learning researcher at NASA Ames, involved in the following research:
- - Developing algorithms for cosmic ray detection to denoise mid-infrared images from the James Webb Space Telescope (JWST).
- - Performing transit detection for TESS and Kepler using machine learning methods.
- - Leading the ExoMiner project, a deep learning-based method that sifted through Kepler data and validated 301 planet candidates.
Background
- An applied machine learning researcher, specializing in the application of ML to astronomy and astrophysics. His research primarily focuses on enhancing the output of exoplanet survey missions, such as Kepler and TESS. He develops ML models to analyze high-quality, large datasets of transit signals, aiming to accelerate the vetting of planet candidates and the validation of exoplanets. Additionally, he has interests in uncertainty quantification, characterization of deep learning models, and model explainability (XAI).
Co-authors