Grant: Discovery Launch Supplement
University of Toronto
$12,500.00
Physics-Informed Machine Learning for Computational Imaging of the Dynamic Universe
EF 2024-25
Résumé
In the last decade, machine learning (ML) has completely revolutionized the fields of computer vision, natural language processing, and robotics; however, it has not been as extensively adopted by other scientific disciplines. This is in part due to: (1) the difficulty of ML models to generalize beyond the training data set and remain open to scientific discovery; and (2) a strong emphasis on intra-disciplinary progress. In recent years, neural representations and physics-informed ML have emerged as powerful interdisciplinary alternatives that infuse ML architectures with physics priors.
The long-term goal of my research program is to develop computational imaging and physics-informed ML for discovery within the physical sciences. A key aspect that differentiates this research program is the interdisciplinary focus on identifying novel inverse imaging problems with the potential to lead to scientific discoveries. Towards this goal, I identify three long term objectives:
(1) Identifying and studying non-linear inverse imaging problems within science (Objective 1)
(2) Studying and developing computational imaging and equation discovery of dynamical systems from sparse and noisy observations (Objective 2)
(3) Advancing and facilitating scientific discovery in astronomy and astrophysics through the application of black hole imaging with the Event Horizon Telescope (Objective 3)
These three distinct research objectives interface synergistically: Research objectives (1, 2) will focus on developing fundamental technologies that exploit neural representations informed by the underlying physics; research objective (3) is supported by my unique position within the Event Horizon Telescope as a coordinator for a collaboration-wide project focused on dynamic imaging of Sagittarius A*, the black hole at the center of the Milky Way.
While the program and, in particular, research objective (3) are geared towards astronomy and astrophysics, they will significantly impact broader areas of science that seek to characterize and study complex dynamical phenomena from observations. Pursuing these objectives will allow us to expand our expertise and further strengthen our existing partnerships with domain experts (e.g., with research groups at NASA and NOAA). This will ultimately enable us to develop new fundamental imaging technologies for earth science, climate science, and other fields".
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
DGECR-2024-00524
Name nom
Levis, Aviad
Department département
Department of Computer Science
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2603
Research subject group en
Robotics
Research subject group fr
Robotique
Research subject en
Computer vision
Research subject fr
Vision artificielle
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Physics-Informed Machine Learning for Computational Imaging of the Dynamic Universe
Keywords
computational imaging; physics informed machine learning; dynamical systems; computational tomography; equation discovery; black hole imaging; event horizon telescope
Application summary
In the last decade, machine learning (ML) has completely revolutionized the fields of computer vision, natural language processing, and robotics; however, it has not been as extensively adopted by other scientific disciplines. This is in part due to: (1) the difficulty of ML models to generalize beyond the training data set and remain open to scientific discovery; and (2) a strong emphasis on intra-disciplinary progress. In recent years, neural representations and physics-informed ML have emerged as powerful interdisciplinary alternatives that infuse ML architectures with physics priors.
The long-term goal of my research program is to develop computational imaging and physics-informed ML for discovery within the physical sciences. A key aspect that differentiates this research program is the interdisciplinary focus on identifying novel inverse imaging problems with the potential to lead to scientific discoveries. Towards this goal, I identify three long term objectives:
(1) Identifying and studying non-linear inverse imaging problems within science (Objective 1)
(2) Studying and developing computational imaging and equation discovery of dynamical systems from sparse and noisy observations (Objective 2)
(3) Advancing and facilitating scientific discovery in astronomy and astrophysics through the application of black hole imaging with the Event Horizon Telescope (Objective 3)
These three distinct research objectives interface synergistically: Research objectives (1, 2) will focus on developing fundamental technologies that exploit neural representations informed by the underlying physics; research objective (3) is supported by my unique position within the Event Horizon Telescope as a coordinator for a collaboration-wide project focused on dynamic imaging of Sagittarius A*, the black hole at the center of the Milky Way.
While the program and, in particular, research objective (3) are geared towards astronomy and astrophysics, they will significantly impact broader areas of science that seek to characterize and study complex dynamical phenomena from observations. Pursuing these objectives will allow us to expand our expertise and further strengthen our existing partnerships with domain experts (e.g., with research groups at NASA and NOAA). This will ultimately enable us to develop new fundamental imaging technologies for earth science, climate science, and other fields".