Learned Exascale Computational Imaging (LEXCI)
Lead Research Organisation:
University College London
Department Name: Mullard Space Science Laboratory
Abstract
The emerging era of exascale computing that will be ushered in by the forthcoming generation of supercomputers will provide both opportunities and challenges. The raw compute power of such high performance computing (HPC) hardware has the potential to revolutionize many areas of science and industry. However, novel computing algorithms and software must be developed to ensure the potential of novel HPC architectures is realized.
Computational imaging, where the goal is to recover images of interest from raw data acquired by some observational instrument, is one of the most widely encountered class of problem in science and industry, with myriad applications across astronomy, medicine, planetary and climate science, computer graphics and virtual reality, geophysics, molecular biology, and beyond.
The rise of exascale computing, coupled with recent advances in instrumentation, is leading to novel and often huge datasets that, in principle, could be imaged for the first time in an interpretable manner at high fidelity. However, to unlock interpretable, high-fidelity imaging of big-data novel methodological approaches, algorithms and software implementations are required -- we will develop precisely these components as part of the Learned EXascale Computational Imaging (LEXCI) project.
Firstly, whereas traditional computational imaging algorithms are based on relatively simple hand-crafted prior models of images, in LEXCI we will learn appropriate image priors and physical instrument simulation models from data, leading to much more accurate representations. Our hybrid techniques will be guided by model-based approaches to ensure effectiveness, efficiency, generalizability and uncertainty quantification. Secondly, we will develop novel algorithmic structures that support highly parallelized and distributed implementations, for deployment across a wide range of modern HPC architectures. Thirdly, we will implement these algorithms in professional research software. The structure of our algorithms will not only allow computations to be distributed across multi-node architectures, but memory and storage requirements also. We will develop a tiered parallelization approach targeting both large-scale distributed-memory parallelization, for distributing work across processors and co-processors, and light-weight data parallelism through vectorization or light-weight threads, for distributing work on processors and co-processors. Our tiered parallelization approach will ensure the software can be used across the full range of modern HPC systems. Combined, these developments will provide a future computing paradigm to help usher in the era of exascale computational imaging.
The resulting computational imaging framework will have widespread application and will be applied to a number of diverse problems as part of the project, including radio interferometric imaging, magnetic resonance imaging, seismic imaging, computer graphics, and beyond. The resulting software will be deployed on the latest HPC computing resources to evaluate their performance and to feed back to the community the computing lessons learned and techniques developed, so as to support the general advance of exascale computing.
Computational imaging, where the goal is to recover images of interest from raw data acquired by some observational instrument, is one of the most widely encountered class of problem in science and industry, with myriad applications across astronomy, medicine, planetary and climate science, computer graphics and virtual reality, geophysics, molecular biology, and beyond.
The rise of exascale computing, coupled with recent advances in instrumentation, is leading to novel and often huge datasets that, in principle, could be imaged for the first time in an interpretable manner at high fidelity. However, to unlock interpretable, high-fidelity imaging of big-data novel methodological approaches, algorithms and software implementations are required -- we will develop precisely these components as part of the Learned EXascale Computational Imaging (LEXCI) project.
Firstly, whereas traditional computational imaging algorithms are based on relatively simple hand-crafted prior models of images, in LEXCI we will learn appropriate image priors and physical instrument simulation models from data, leading to much more accurate representations. Our hybrid techniques will be guided by model-based approaches to ensure effectiveness, efficiency, generalizability and uncertainty quantification. Secondly, we will develop novel algorithmic structures that support highly parallelized and distributed implementations, for deployment across a wide range of modern HPC architectures. Thirdly, we will implement these algorithms in professional research software. The structure of our algorithms will not only allow computations to be distributed across multi-node architectures, but memory and storage requirements also. We will develop a tiered parallelization approach targeting both large-scale distributed-memory parallelization, for distributing work across processors and co-processors, and light-weight data parallelism through vectorization or light-weight threads, for distributing work on processors and co-processors. Our tiered parallelization approach will ensure the software can be used across the full range of modern HPC systems. Combined, these developments will provide a future computing paradigm to help usher in the era of exascale computational imaging.
The resulting computational imaging framework will have widespread application and will be applied to a number of diverse problems as part of the project, including radio interferometric imaging, magnetic resonance imaging, seismic imaging, computer graphics, and beyond. The resulting software will be deployed on the latest HPC computing resources to evaluate their performance and to feed back to the community the computing lessons learned and techniques developed, so as to support the general advance of exascale computing.
Publications
Wallis C
(2022)
Mapping dark matter on the celestial sphere with weak gravitational lensing
in Monthly Notices of the Royal Astronomical Society
Spurio Mancini A
(2023)
Bayesian model comparison for simulation-based inference
in RAS Techniques and Instruments
Spurio Mancini A.
(2022)
Bayesian model comparison for simulation-based inference
in arXiv e-prints
Price Matthew A.
(2023)
Differentiable and accelerated spherical harmonic and Wigner transforms
in arXiv e-prints
Price Matthew A.
(2024)
Differentiable and accelerated wavelet transforms on the sphere and ball
in arXiv e-prints
Price M
(2023)
Fast emulation of anisotropies induced in the cosmic microwave background by cosmic strings
in The Open Journal of Astrophysics
Polanska Alicja
(2023)
Learned harmonic mean estimation of the marginal likelihood with normalizing flows
in arXiv e-prints
Pan B
(2023)
On Learning the Invisible in Photoacoustic Tomography with Flat Directionally Sensitive Detector
in SIAM Journal on Imaging Sciences
Ocampo Jeremy
(2022)
Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions
in arXiv e-prints
Munshi D.
(2021)
Position-Dependent Correlation Function of Weak Lensing Convergence
in arXiv e-prints
Munshi D
(2022)
Weak lensing trispectrum and Kurt-spectra
in Journal of Cosmology and Astroparticle Physics
Munshi D
(2022)
On weak lensing response functions
in Journal of Cosmology and Astroparticle Physics
Munshi D
(2023)
Position-dependent correlation function of weak-lensing convergence
in Physical Review D
McEwen Jason D.
(2023)
Proximal nested sampling with data-driven priors for physical scientists
in arXiv e-prints
McEwen Jason D.
(2021)
Machine learning assisted Bayesian model comparison: learnt harmonic mean estimator
in arXiv e-prints
Mars Matthijs
(2023)
Learned Interferometric Imaging for the SPIDER Instrument
in arXiv e-prints
Mars M
(2023)
Learned interferometric imaging for the SPIDER instrument
in RAS Techniques and Instruments
Marignier Augustin
(2022)
Sparse Bayesian mass-mapping using trans-dimensional MCMC
in arXiv e-prints
Marignier Augustin
(2021)
Posterior sampling for inverse imaging problems on the sphere in seismology and cosmology
in arXiv e-prints
Marignier A
(2023)
Posterior sampling for inverse imaging problems on the sphere in seismology and cosmology
in RAS Techniques and Instruments
Marignier A
(2023)
Sparse Bayesian mass-mapping using trans-dimensional MCMC
in The Open Journal of Astrophysics
Liaudat
(2023)
Scalable Bayesian uncertainty quantification with data-driven priors for radio interferometric imaging
in arXiv e-prints
Cai Xiaohao
(2021)
Proximal nested sampling for high-dimensional Bayesian model selection
in arXiv e-prints
Cai X
(2022)
Proximal nested sampling for high-dimensional Bayesian model selection
in Statistics and Computing