Daniel Muthukrishna received his PhD in Astronomy from the University of Cambridge in 2021. His research has primarily been focused on applying deep neural networks and Bayesian statistics to time-series astronomical data.
He is the developer of RAPID (https://astrorapid.readthedocs.io), a neural-network based classification tool to photometrically classify supernovae and other extragalactic transients. RAPID was developed in preparation for the LSST (Legacy Survey of Space and Time), but is currently being applied to ZTF and PanSTARRS. Daniel has also worked with the Dark Energy Survey (DES), and built a spectroscopic classification tool, DASH (https://astrodash.readthedocs.io), that is currently being used to rapidly classify supernova spectra. His most recent work is working to prepare for the deluge of time-series data from upcoming surveys such as LSST. He has developed a time-series anomaly detection framework that uses a Bayesian modelling process to find new objects and ‘needles in the haystack’ from ZTF.
Daniel has just started work with the TESS (Transiting Exoplanet Survey Satellite) team to help detect interesting transients in the high cadence data-stream.
Selected Papers
- D. Muthukrishna, G. Narayan, K. Mandel, R. Biswas, and R. Hlozek, RAPID: Early Classification of Explosive Transients using Deep Learning, PASP. (2019).
- D. Muthukrishna, D. Parkinson and B. Tucker, DASH: Deep Learning for the Automated Spectral Classification of Supernovae and their Hosts, ApJ. (2019).
- D. Muthukrishna and D. Parkinson, A cosmographic analysis of the transition to acceleration using SN-Ia and BAO, J. Cosmol. Astropart. Phys (2016).