Hybrid Physical-Deep Learning Model for Astronomical Inverse Problems
arXiv:1912.03980
Abstract
We present a Bayesian machine learning architecture that combines a physically motivated parametrization and an analytic error model for the likelihood with a deep generative model providing a powerful data-driven prior for complex signals. This combination yields an interpretable and differentiable generative model, allows the incorporation of prior knowledge, and can be utilized for observations with different data quality without having to retrain the deep network. We demonstrate our approach with an example of astronomical source separation in current imaging data, yielding a physical and interpretable model of astronomical scenes.
8 pages, accepted submission to the NeurIPS 2019 Machine Learning and the Physical Sciences Workshop
References in corpus (6)
- Pixel Recurrent Neural Networks
- Euclid Definition Study Report
- The COSMOS Survey: Hubble Space Telescope / Advanced Camera for Surveys (HST/ACS) Observations and Data Processing
- Recurrent Inference Machines for Solving Inverse Problems
- SCARLET: Source separation in multi-band images by Constrained Matrix Factorization
- Deblending galaxy superpositions with branched generative adversarial networks
Cited by in corpus (5)
- Machine learning and the physical sciences
- Super-resolution emulator of cosmological simulations using deep physical models
- Dynamical mass inference of galaxy clusters with neural flows
- Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modeling
- Solving Inverse Problems for Spectral Energy Distributions with Deep Generative Networks