Speakers
Christina Reissel
(Massachusetts Inst. of Technology (US))
Maira Khan
(Fermi National Accelerator Laboratory)
Description
We investigate the application of state space models (SSMs) to a diverse set of scientific time series tasks. In particular, we benchmark the performance of SSMs against a set of baseline neural networks across three domains: magnet quench prediction, gravitational wave signal classification (LIGO), and neural phase estimation. Our analysis evaluates both computational efficiency—quantified by the number of mathematical operations—and task-specific performance metrics. Results suggest that SSMs offer a favorable trade-off between accuracy and computational efficiency, making them a possible alternative to conventional deep learning models in scientific settings.
Authors
Christina Reissel
(Massachusetts Inst. of Technology (US))
Maira Khan
(Fermi National Accelerator Laboratory)
Co-authors
Abhijith Gandrakota
(Fermi National Accelerator Lab. (US))
Ahmed Enis Cetin
(UIUC)
Emadeldeen Hamdan
(University of Illinois Chicago)
Jennifer Ngadiuba
(FNAL)
Liam Sheldon
Mengke Zhang
Nhan Tran
(Fermi National Accelerator Lab. (US))
Philip Coleman Harris
(Massachusetts Inst. of Technology (US))
Tao Wei
(Clemson University)
