TY - BOOK
T1 - Field Work Proposal ERKJ358: Black-box training for scientific machine learning models (Final Report)
AU - Zhang, Guannan
AU - Bement, Matt
AU - Tran, Hoang
PY - 2022
Y1 - 2022
N2 - The overarching goal of this project is to develop a scalable black-box training capability for scientific machine learning (SciML) problems that are non-trainable with existing automatic differentiation (AD)-based algorithms. AD assumes that a loss function can be decomposed into a sequence of elementary operations whose derivatives are known. This assumption is violated when the loss function includes a black-box physical model (e.g., a legacy simulator). The current strategy, converting a black-box simulator to an AD-enabled code via differential programming, is inflexible and time-, labor-consuming. Thus, black-box optimization is a main workhorse for training SciML models, e.g., in scientific reinforcement learning, hyper-parameter fine tuning, designing SciML models with adversarial robustness, etc.
AB - The overarching goal of this project is to develop a scalable black-box training capability for scientific machine learning (SciML) problems that are non-trainable with existing automatic differentiation (AD)-based algorithms. AD assumes that a loss function can be decomposed into a sequence of elementary operations whose derivatives are known. This assumption is violated when the loss function includes a black-box physical model (e.g., a legacy simulator). The current strategy, converting a black-box simulator to an AD-enabled code via differential programming, is inflexible and time-, labor-consuming. Thus, black-box optimization is a main workhorse for training SciML models, e.g., in scientific reinforcement learning, hyper-parameter fine tuning, designing SciML models with adversarial robustness, etc.
KW - 97 MATHEMATICS AND COMPUTING
U2 - 10.2172/1905375
DO - 10.2172/1905375
M3 - Commissioned report
BT - Field Work Proposal ERKJ358: Black-box training for scientific machine learning models (Final Report)
CY - United States
ER -