MS166A Advancing SCiML Surrogates via Numerical Methods and Vice Versa I
MS Corresponding Organizer: Dr. Alena Kopanicakova (Brown University)
Chaired by:
Dr. Alexander Heinlein (TU Delft , Netherlands) , Dr. Alena Kopanicakova (Brown University , United States)
Dr. Alexander Heinlein (TU Delft , Netherlands) , Dr. Alena Kopanicakova (Brown University , United States)
Scheduled presentations:
-
Learning Adaptive Coarse Basis Functions of FETI-DP
A. Klawonn, M. Lanser, J. Weber* -
Training Deep Neural Networks Using Multilevel and Domain Decomposition Methods
A. Kopanicakova* -
Iterative algorithms for partitioned neural network approximation to partial differential equations
H. Kim*, H. Yang, D. Jang, K. Kim, Y. Jeon -
Towards ExaScale Training: Domain Decomposition Methods for the Efficient Training of Neural Networks
R. Krause*, S. Cruz, A. Kopanicakova, H. Kothari, K. Trotti -
Algorithm Switching for Multiobjective Derivative Free Optimization
Z. Li, A. Kannan* -
Data-driven surrogate models for an efficient numerical homogenization of open-porous biopolymer aerogels
A. Klawonn, M. Lanser, L. Mager*, A. Rege
