The NSF-Simons AI Institute for the Sky (SkAI Institute) is one of the National Artificial Intelligence Research Institutes funded by the National Science Foundation (NSF) and Simons Foundation, which will accelerate Astro-AI research and help educate a diverse Astro-AI workforce. CAPS is one of the lead partners of the SkAI Institute.
CAPS Work on the SkAI Institute Project:
CAPS Personnel Holding Leadership Roles on the SkAI Institute Project:
Jay Alameda: Technical Advisor on the SkAI Operations Team; Member of the SkAI Conference 2025 and 2026 (Open SkAI 2025 and Open SkAI 2026) Local Organizing Committees
Armgard Haken: Project Development and Data Manager on the SkAI Operations Team; Member of the SkAI Conference 2025 (Open SkAI 2025) Local Organizing Committee
Scott Koranda: Research Software Engineer
Matthew Krafczyk: Research Software Engineer
Jennifer Li: Research Software Engineer; Member of the SkAI Conference 2025 (Open SKAI 2025) Scientific Organizing Committee
Billy Moses: Member of the SkAI Conference 2025 (Open SkAI 2025) Scientific Organizing Committee
Gautham Narayan: Co-Principal Investigator; Deputy Director for Research (Astro); Member of the SkAI Conference 2025 (Open SkAI 2025) Scientific Organizing Committee
Matthew Turk: Activities and Programming Co-Chair for Community Software, Data, & Hardware Resources; Member of the SkAI Conference 2025 (Open SkAI 2025) Local Organizing Committee
Joaquin Vieira: Co-Chair of SkAI Institute Pillar 3 (Learning-Based Astrophysical Survey and Instrument Design)
Amanda Wasserman (former CAPS Graduate Student Fellow): Assistant Director for Research Operations on the SkAI Operations Team; Member of the SkAI Conference 2025 and 2026 (Open SkAI 2025 and Open SkAI 2026) Local Organizing Committees
Nicolas Yunes: Activities and Programming Co-Chair for the Public Outreach & High School Program
Current CAPS Personnel on the SkAI Institute Project:
Aadya Agrawal (CAPS-SkAI Graduate Student Fellow)
Jay Alameda (CAPS Senior Technical Program Manager)
Fabián Araneda-Baltierra (CAPS Visiting Research Scientist)
Matias Carrasco Kind (CAPS affiliated faculty)
Siegfried Eggl (CAPS affiliated faculty)
Bill Gropp (NCSA Immediate Former Director)
Armgard Haken (CAPS Research Coordinator)
Scott Koranda (CAPS Senior Research Scientist)
Matthew Krafczyk (CAPS Research Scientist)
Jennifer Li (CAPS Research Scientist)
Santiago Loane (CAPS-SkAI Graduate Student Fellow)
Billy Moses (CAPS affiliated faculty)
Gautham Narayan (CAPS Deputy Director)
Matthew Turk (CAPS affiliated faculty)
Joaquin Vieira (CAPS Director)
Nicolas Yunes (CAPS affiliated faculty)
SkAI-Funded Projects Involving CAPS Personnel:
Title: SELDON: A foundation AI model to infer the physics of transients from cosmic surveys
Co-Lead PI: Gautham Narayan
Co-PI: Matt Krafczyk
Collaborators: Jennifer Li, Amanda Wasserman (former CAPS Graduate Student Fellow), Haille Perkins (former CAPS Graduate Student Fellow)
Project Summary: This project aims to build a foundational AI model that would form a core part of a community pipeline for Rubin to identify (including anomalies) and characterize transients and set priorities for follow-up observations. The foundation model will be based on a new custom variational autoencoder (VAE) architecture, optimized for the sparse, multimodal and irregular data, and both observational data and emulated data from (physical) transient models. In year 2, the VAE will be replaced by a transformer.
Title: Automating Bayesian inference of millimeter source association
Lead PIs: Joaquin Vieira, Billy Moses
Collaborators: Jennifer Li, Melanie Archipley (former CAPS Graduate Student Fellow), Kedar Phadke (former CAPS Graduate Student Fellow)
Project Summary: This project aims to develop an automated Bayesian inference tool for millimeter source association [cross-matching South Pole Telescope (SPT) sources with other surveys]. Correct associations to observations in different wavelengths are necessary to reconstruct spectral energy distributions (SEDs) and infer properties and redshifts of astrophysical objects. Bayesian inference is one of the methods to do this, but due to large numbers of objects that will be available from the current and future surveys, faster methods are necessary. This project aims to use machine learning (ML) to automate Bayesian inference, focusing on advancements in differentiable and probabilistic programming. Unlike existing differential programming problems, this project will incorporate the use of discrete variables, which are not handled well by existing automatic differentiation (AD) frameworks. The team plans to build a Zooniverse project to confirm source associations via visual inspection by volunteers.
Title: AI-accelerated simulations with multiscale astrophysics using physics-based deep learning
Co-PI: Billy Moses
Collaborator: Matt Krafczyk
Project Summary: This project has the following objectives: (1) model stellar structure and evolution, (2) combine learned components (trained to emulate simulation) with numeric methods, (3) order-of-magnitude speedups, (4) enable modeling large populations of (single and binary) stars, and (5) incorporate uncertainty quantification. These objectives will be met using (1) physics-informed neural networks (PINNs) and physics-informed neural operators (PINOs), (2) hybrid combination of physics-based deep learning (PBDL) with classical methods, and (3) active learning/mesh refinement.
Title: Intelligent scheduling for astronomical surveys
Collaborator: Gautham Narayan
Project Summary: This project has the following objectives: (1) develop reinforcement-learning (RL) algorithms for dynamic, real-time astronomical observation scheduling with deferred, multi-objective rewards; and (2) use active learning and weak-to-strong supervision to optimize target selection for population-level parameter inference. These objectives will be met by (1) leveraging state-of-the-art AI techniques, including RL, active learning, and conformal prediction; and (2) using ∼500k observations from the Dark Energy Camera (DECam) for model training and validation.
Title: TEMPI: A world model for the time-domain sky
Co-Lead PI: Gautham Narayan
Project Summary: TEMPI represents a new flagship project within SkAI for the Legacy Survey of Space and Time (LSST)/Rubin era: to move time-domain astrophysics from task-specific machine-learning pipelines toward a unified, physics-aware world model of the dynamic sky. At present, SkAI has supported two efforts (SELDON, focusing on transients, and StarEmbed, focusing on variable stars). Now we will extend beyond these projects to work with real data from the U.S. National Science Foundation-U.S. Department of Energy (NSF-DOE) Vera C. Rubin Observatory (LSST alerts started in February 24, 2026), National Aeronautics and Space Administration (NASA) Nancy Grace Roman Space Telescope (launching August–September, 2026), and European Space Agency (ESA) Euclid Space Telescope, enabling them to work with more complex data modalities, and to interface with real-time prioritization and follow-up services with agentic artificial intelligence (AI).
Title: Radio antenna design through learning latent space characterizations of 3D structures
Co-Lead PI: Matt Krafczyk
Project Summary: This project brings AI-enabled design methods to a central obstacle in early-Universe cosmology: measuring the sky-averaged 21-cm signal from the cosmological Dark Ages requires antennas whose frequency-dependent response does not imprint spectral structure that mimics the signal itself. We treat antenna design as a foundational AI problem of learning representations of, and optimizing, three-dimensional objects subject to hard physical and manufacturing constraints. The resulting methods will be broadly applicable to low-frequency cosmology, radio astronomy, and particle astrophysics, and will be validated experimentally through antennas we fabricate and characterize.
