Matthew Craigie
JPL Postdoc
Biography
Matt uses machine learning methods to drive scientific discovery in the field of cosmology. He completed his undergraduate studies and PhD at the University of Queensland in Brisbane, Australia, where his thesis focused on developing machine learning models that overcome the unique challenges in cosmology datasets. Matt joined JPL as a JPL Postdoc in 2025, where he uses deep learning approaches to better understand the tidal interactions between galaxies, unlocking insights into galaxy evolution and the history of the universe.
Education
- Ph. D., The University of Queensland, Australia (2025)
- B. Adv. Sc. (Hons) with a Major in Physics, The University of Queensland, Australia (2021)
Professional Experience
- JPL Postdoc (2025)
Research Interests
- Weak Lensing Cosmology
- Deep Learning: Physics-based architectures and model interpretability
- Galaxy Clustering and Evolution
Selected Publications
- M. Craigie et al. Learning Intrinsic Alignments from Local Galaxy Environments. Submitted to Phys. Rev. D. (2025)
- M. Craigie et al. Learning Balanced Field Summaries of the Large-Scale Structure with the Neural Field Scattering Transform. Submitted to Phys. Rev. D. (2025)
- M. Craigie et al. Unsupervised Searches for Cosmological Parity Violation: Improving detection power with the Neural Field Scattering Transform. Phys. Rev. D. (2024)
- P. Taylor, M. Craigie & Y.-S. Ting. Unsupervised searches for cosmological parity violation: An investigation with convolutional neural networks. Phys. Rev. D. (2024)
- Y. Shendryk, M. Craigie et al. A Satellite-Based Methodology for Harvest Date Detection and Yield Predic-tion in Sugarcane. IGARSS (2020)