Machine Learning in Survey Cosmology

The NASA Postdoctoral Program (NPP) offers unique research opportunities to highly-talented scientists to engage in ongoing NASA research projects. This specific postdoctoral fellowship focuses on applying deep learning techniques to cosmology missions in the big data era of astronomy. The research will utilize architectures based on convolutional neural networks (GANs, deep image prior, diffusion probabilistic models) for applications such as image generation, data/image translation, and super resolution to enhance astronomical measurements from combined datasets. The fellowship aims to boost images in depth and resolution to increase usable sources for weak lensing, reduce systematic errors, and put tighter constraints on cosmological parameters. The successful candidate will work with data from current ground- and space-based surveys (HSC, KiDS+Viking, DES, HST) to quantify gains for upcoming missions (Rubin, Euclid, Roman, SPherex) with JPL, Caltech, and IPAC contributions. The fellow will work closely with Dr. Eric Huff, Dr. Shoubaneh Hemmati, and other members of the cosmology team at JPL. These competitive fellowships are designed to advance NASA's missions in space science, Earth science, aeronautics, space operations, exploration systems, and astrobiology.