Machine Learning and HST/WFC3: An Update for 2023

Dauphin, Frederick; Medina, Jennifer; Montes Quiles, Mireia; Bajaj, Varun; Easmin, Nilufar; McCullough, Peter
Referencia bibliográfica

American Astronomical Society Meeting Abstracts

Fecha de publicación:
1
2023
Número de autores
6
Número de autores del IAC
1
Número de citas
0
Número de citas referidas
0
Descripción
Hubble's Wide Field Camera 3 (WFC3) is the workhorse instrument for HST, providing direct (staring) and scanning modes using filters and grisms, covering from the near-IR to near-UV. Installed during the most recent HST servicing mission in 2009, WFC3 has logged almost 300,000 observations resulting in exciting scientific discoveries over the past 14 years. With this abundance of data and growing accessibility to artificial intelligence, we utilize machine learning for detecting anomalies in our images, specifically WFC3/IR Blobs and WFC3/UVIS Figure-8 Ghosts. We discuss our models' performances and future projects, such as utilizing dimensionality reduction and clustering to further explore the WFC3 database. In addition, we highlight items of particular interest to proposers in Cycle 31 and observers with data in hand.