.png){kind=icon}
Return
Active operator learning with predictive uncertainty quantification for partial differential equations
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jcp.2026.114791.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Deep neural operators are equipped with predictive uncertainties to accurately estimate errors. • Uncertainty predictions generalize to out-of-distribution data and function spaces. • We introduce an optimized inference implementation suited for real-time applications. • Predictive uncertainties are used to accelerate active learning in function spaces.
Keywords:
Deep neural operators
Predictive uncertainty
Active learning
Partial differential equations
Uncertainty quantification
Journal
IF:
3.8
Papers:
1.6W
Citations:
7.4W
Organization
Cited Papers
ShareSaveShareSave
ShareSaveShareSaveLearning the solution operator of parametric partial differential equations with physics-informed DeepONets.png)
SCIENCE ADVANCES
OAAI
SCIENCE ADVANCESIF12.5
OAAIShareSave
View more