AstroAI Lunch Talks - March 10, 2025 - ChangHoon Hahn
10 Mar 2025 - Joshua Wing
The video can be found here: https://www.youtube.com/watch?v=dtgCV-xf4i8
Speaker: ChangHoon Hahn (Princeton University & University of Arizona)
Title: Reconstructing Quasar Spectra and Measuring the Lyα Forest with Spectrum Autoencoders
Abstract: Quasar spectra carry the imprint of foreground intergalactic medium (IGM) through absorption systems. In particular, absorption caused by neutral hydrogen gas, the “Lyα forest,” is a key spectroscopic tracer for cosmological analyses used to measure cosmic expansion and test physics beyond the standard model. Despite their importance, current methods for measuring Lyα absorption require making strong assumptions on the shape of the intrinsic quasar continuum that bias Ly α analyses. In this talk, I will present SpenderQ, a ML-based approach for reconstructing intrinsic quasar spectra and measuring the Lyα forest directly from observations. SpenderQ leverages a spectrum autoencoder to learn a compact and redshift-invariant latent representation of quasar spectra, combined with an iterative procedure to identify and mask absorption regions in the spectra. I will demonstrate that SpenderQ accurately reconstructs the true intrinsic quasar spectra, including the broad Lyβ, Lyα, SiIV, CIV, and CIII emission lines, with percent-level precision. I will also show how SpenderQ provides physically meaningful latent encoding enables key downstream tasks such as classification and anomaly detection. Finally, I will discuss how SpenderQ will be applied to observations to provide unbiased Lyα forest measurements in cosmological, quasar, and IGM studies.