I am currently mentoring two undergraduate students on machine learning research in astronomy.
- Siddhant Solanki is a senior student in physics at UC Santa Barbara. We are exploring the application of data augmentation in improving light curve classification performance. Classification tasks face two main challenges: (1) labeled datasets require extensive human attention to create; (2) class distributions in real-life datasets are highly imbalanced. With most samples coming from a few highly populated classes, imbalanced datasets are prone to poor predictive accuracy in the often more interesting minority classes. Our goal is to tackle these challenges by exploiting the representative power of the feature space to increase the effective size and diversity of the datasets. Siddhant is applying for grad school in astronomy this semester, look out for his application!
- Abhay Agarwal is a junior student in applied math and physics at UC Berkeley. We are applying convolutional neural networks to analyze the statistics of isothermal turbulent gas clouds. Recent molecular line observations by ALMA have begun to probe the properties of massive star-forming clouds in nearby galaxies. The goal is to develop a pipeline to help constrain the kinematics of star-forming gas observed by current and future telescopes.
I enjoy teaching and sharing the latest news in astronomy with my students. During grad school, I was a teaching assistant in 7 different undergraduate courses (some for multiple semesters). They span from first-year introductory to advanced astronomy courses:
- Introduction to Astronomy
- Popular Astronomy
- Search for Extraterrestrial Life and Intelligence
- Galaxies and the Universe
- Births of Stars and Planets
- Stellar Astronomy