Projects

1. Generative modelling with neural-network based Conditional Normalizing Flows to remove scattered light artefacts in NASA’s Transiting Exoplanet Survey Satellite’s Full-frame images
2. Semi-supervised and self-supervised learning for modelling time-series data
3. Anomaly detection in real-time
4. Real-time prediction of the age of astrophysical transients
5. Exoplanet light curve classification

Anomaly Detection in real-time for astronomical transients

• Using Bayesian recurrent neural networks to encode astrophysical transients. Predicting anomalies with predictive modelling or Isolation Forests
• Paper on “Real-Time Detection of Anomlaies in Large-Scale Transient Surveys” in the Monthly Notices of the Astronomical Society
• Paper in the Astronomy workshop at the Neurips 2021 “Real-time Detection of Anomalies in Multivariate Time Series of Astronomical Data“

COVID-19 Modelling

• Developed App to track and compare COVID-19 across countries:
  • https://covid19-cases-and-deaths.herokuapp.com/
• Developed App with Nick Taylor (Plant Scientist PhD, University of Cambridge) to help public understanding of COVID-19 modeling and control measures
  • https://lowhighcovid.herokuapp.com/
  • Media reports on the project
  •  Interview on BlueSci Podcast

RAPID: Early classification of photometric transients for LSST and ZTF

• Paper on “RAPID: Early Classification of Explosive Transients using Deep Learning” (Muthukrishna et al., 2019b)
• Open source software: https://astrorapid.readthedocs.io
• Try pip install astrorapid  to get started!
• Finalist in the ongoing “Best Astrostatistics Student Paper Award” by the ASA Astrostatistics Interest Group.
  

Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC)

• Working with the PLAsTiCC collaboration to develop a community-wide transient classification challenge on Kaggle (kaggle.com/c/PLAsTiCC-2018).
• Paper on “Models and Simulations for the Photometric LSST Astronomical Time Series Classification Challenge (PLAsTiCC)” (Kessler, 2019b).
• Paper on “The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Selection of a performance metric for classification probabilities balancing diverse science goals” (Malz et al. 2019).
• Note on “The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Data set” (The PLAsTiCC Team, 2018).

DASH: Deep learning for the automated classification of supernovae

• Paper on “DASH: Deep Learning for the Automated Spectral Classification of Supernovae and their Hosts” (Muthukrishna et al., 2019a)
• Thesis
• Documentation
• Github
• Try pip install astrodashto get started!
• My thesis has won several prizes for its novel approach.
  • Bok Prize Highly Commended (Awarded to the top three astronomy-related theses from all Honours/Masters students in Australia)
  • IEEE Student Thesis Prize (Awarded to the top ranked Information Technology and Electrical Engineering thesis from all undergraduate students in Queensland)
  • GBST Best Software Project (Awarded to the top ranked software-related thesis at the University of Queensland)

Dark Energy Survey

• Paper on “OzDES multifibre spectroscopy for the Dark Energy Survey: Three year results and first data release” (Childress et al., 2017)
• Paper on “First Cosmology Results using Type Ia Supernovae from the Dark Energy Survey: Constraints on Cosmological Parameters” (DES Collaboration, 2019)
• Paper on “First Cosmology Results Using Type Ia Supernovae From the Dark Energy Survey: Survey Overview and Supernova Spectroscopy” (D’Andrea, 2019)
• Paper on “First Cosmology Results using Type Ia Supernova from the Dark Energy Survey: Simulations to Correct Supernova Distance Biases” (Kessler, 2019a)
• Paper on “First Cosmology Results Using Type Ia Supernovae From the Dark Energy Survey: Analysis, Systematic Uncertainties, and Validation” (Brout, 2019)

SkyMapper Transient Survey

• Working with the SkyMapper team to hunt for new supernovae! We are using the 1.3m telescope at Siding Spring Observatory to study supernovae and other transient objects across ~1000 square degrees of the sky.
• We recently launched the citizen-science project to search for Planet 9 (launched on BBC’s Stargazing Live TV event). The supernova citizen-science search was also recently launched and is live here.
• The SkyMapper Team includes Brian Schmidt, Anais Möller, Brad Tucker, Ashley Ruiter, Ivo Seitenzahl, Seo-Won Chang, Bonnie Zhang, Fiona Panther, Daniel Muthukrishna, Natalia E. Sommer, Ryan Ridden-Harper, Patrick Armstrong, Nataliea Lowson, Richard Scalzo, Fang Yuan, Mike Childress, Chris Onken, Chris Wolf.

Model-independent cosmology with SNIa and BAO data

• Paper on “A cosmographic analysis of the transition to acceleration using SN-Ia and BAO” (Muthukrishna & Parkinson, 2016)

Modelling emission line profiles with multiple Gaussian components

• V. Firpo, D. Muthukrishna, F. Campuzano-Castro, S. Torres-Flores, G. Bosch and G. Hagele, Violent star-forming processes in interacting galaxies, (2018). (in prep)
• Paper on “Estudio en galaxias BCD: Mrk 600 y IIZw 33” (Campuzano Castro, 2018)
• GitHub

Design of a portable digital oscilloscope

• With Benjamin Jarrett and Ovindu Manawaduge