NASA, Jet Propulsion Laboratory, California Institute of Technology
Colloquium Tea held at 4:00 pm in 154 Hurley Hall
Kernel Flow Emulation for NASA's Surface Biology and Geology Mission.
NASA's new Surface Biology and Geology (SBG) mission will launch in late 2026 and carry a hyperspectral imager to observe Earth's surface at high resolution (~30 meter) in the visible and thermal regions of the electromagnetic spectrum. Daily data volume is expected to be 2.5 to 5 petabytes. The mission's science objectives include understanding active surface changes, snow and ice accumulation, hazard risks, changing land use, plant physiology, and terrestrial and aquatic ecosystems. To meet these objectives, geophysical properties of Earth's surface must be inferred from observed spectra. Spectra are related to surface states via physical forward models embedded within inference algorithms. These forward models are computationally demanding, and will require emulation in order to keep up with data flow. In this talk we introduce a forward model emulator for SBG using a new method for fitting covariance parameters of Gaussian Processes, called Kernel Flows (KF; Owhadi and Yoo, 2019). KF uses mini-batch stochastic gradient descent and cross-validation to achieve robust estimates in a computationally efficient manner. KF has deep connections to neural networks when viewed through the lens of decision theory. The innovation of this work is in the computational implementation of the algorithm, and its application in the remote sensing context.
Bio: Dr. Amy Braverman is a Senior Research Scientist at the Jet Propulsion Laboratory, California Institute of Technology, in Pasadena, CA. She is the Technical Group Lead for Statistical Methods and Applications in the Uncertainty Quantification and Statistical Analysis Group of the Instrument Operations and Science Data Systems Section. After graduating from Swarthmore College in 1982 with a B.A. in Economics, Dr. Braverman worked for nearly a decade in litigation support consulting. She returned to graduate school at UCLA in the early 1990’s where she earned an M.A. in Mathematics and Ph.D. in Statistics. She began her career as a post-doc at JPL in 1999 and has been with the Lab ever since. Dr. Braverman’s early work was in the use of data compression methods for analysis of massive data sets. As her career advanced she has worked in spatial and spatio-temporal statistics, statistical methods for the evaluation of climate models, and most recently in Uncertainty Quantification. She has been at the forefront of JPL’s efforts to bring rigorous UQ to the derivation of geophysical information from remote sensing observations collected by NASA and JPL instruments. Dr. Braverman finds special satisfaction in mentoring post-docs and young researchers to build capability in Statistics at JPL, and in collaborating with academic colleagues to connect their research, and that of their graduate students, to JPL and NASA problems.