SIFARI: Self-Supervised Interferometric Fitting for Astronomical Radio Imaging
Abstract
Radio-interferometric images are reconstructed from sparsely sampled visibilities, and CLEAN-based imaging can struggle with spatial filtering, complex morphologies, and uncertainty quantification. Alternative methods that fit visibilities directly can address some of these limitations but often require manual choices of image priors and model hyperparameters. We present SIFARI (Self-Supervised Interferometric Fitting for Astronomical Radio Imaging), a self-supervised neural network workflow that represents sky brightness as a continuous function of position and fits measured visibilities without an external image training set or explicit spatial regularizer. An empirical rule sets the Fourier feature scale from the visibilities before training, controlling how readily the network fits fine structure. Sampling network weights with Stochastic Weight Averaging-Gaussian (SWAG) gives approximate brightness uncertainty estimates, which we combine with a thermal-noise floor to construct spatially resolved signal-to-noise maps. In synthetic ALMA tests, SIFARI yields an effective point-source response about eight times narrower than the natural-weighting CLEAN restoring beam and recovers more extended flux than CLEAN when short baselines are missing. It also achieves higher image fidelity than the restored CLEAN images in all three morphology benchmarks. Applied to ALMA observations of PDS 70, SIFARI recovers the bright outer ring together with faint compact emission in the central cavity. For long-baseline-only WISPIT 2 data, SIFARI supplies a sky model for phase self-calibration where the CLEAN model is inadequate. The restored, self-calibrated SIFARI image has approximately 30% lower RMS noise than the CLEAN image made from the original visibilities without self-calibration.
Related
- I2C@2024: an interferometric imaging contest in 2024
- LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging
- Interferometric Survey of Stellar Parameters: Limb darkening study with the Pipeline for Interferometric Measurements of Stars
- IViS: Interferometric Visibility-domain inversion Software. A GPU-accelerated Python framework for joint deconvolution with ASKAP
- Interferometric Survey of Stellar Parameters: Limb darkening study with the "Pipeline for Interferometric Measurements of Stars
BibTeX
@misc{mao2026sifari,
title = {{SIFARI: Self-Supervised Interferometric Fitting for Astronomical Radio Imaging}},
author = {Shunyuan Mao and Andrea Isella and Paris Perdikaris and Li-Ta Lo and Hui Li},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.35966},
eprint = {2609.35966},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.35966},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - SIFARI: Self-Supervised Interferometric Fitting for Astronomical Radio Imaging
AU - Shunyuan Mao
AU - Andrea Isella
AU - Paris Perdikaris
AU - Li-Ta Lo
AU - Hui Li
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.35966
UR - https://arxiv.org/abs/2609.35966
AB - Radio-interferometric images are reconstructed from sparsely sampled visibilities, and CLEAN-based imaging can struggle with spatial filtering, complex morphologies, and uncertainty quantification. Alternative methods that fit visibilities directly can address some of these limitations but often require manual choices of image priors and model hyperparameters. We present SIFARI (Self-Supervised Interferometric Fitting for Astronomical Radio Imaging), a self-supervised neural network workflow that represents sky brightness as a continuous function of position and fits measured visibilities without an external image training set or explicit spatial regularizer. An empirical rule sets the Fourier feature scale from the visibilities before training, controlling how readily the network fits fine structure. Sampling network weights with Stochastic Weight Averaging-Gaussian (SWAG) gives approximate brightness uncertainty estimates, which we combine with a thermal-noise floor to construct spatially resolved signal-to-noise maps. In synthetic ALMA tests, SIFARI yields an effective point-source response about eight times narrower than the natural-weighting CLEAN restoring beam and recovers more extended flux than CLEAN when short baselines are missing. It also achieves higher image fidelity than the restored CLEAN images in all three morphology benchmarks. Applied to ALMA observations of PDS 70, SIFARI recovers the bright outer ring together with faint compact emission in the central cavity. For long-baseline-only WISPIT 2 data, SIFARI supplies a sky model for phase self-calibration where the CLEAN model is inadequate. The restored, self-calibrated SIFARI image has approximately 30% lower RMS noise than the CLEAN image made from the original visibilities without self-calibration.
ER -