MO-SP.1P: Application of Machine/Deep Learning and Uncertainty Quantification Techniques in Computational Electromagnetics |
| Session Type: Special Session Oral |
| Time: Monday, July 8, 13:20 - 17:00 |
| Location: Grand Ballroom C |
| Session Chairs: Luis Gomez, Duke University School of Medicine, Abdulkadir Yucel, Nanyang Technological University, Cynthia Furse, University of Utah and Costas Sarris, University of Toronto
|
| |
| 13:20 - 13:40 |
| MO-SP.1P.1: GENERALIZATION CAPABILITIES OF DEEP LEARNING SCHEMES IN SOLVING INVERSE SCATTERING PROBLEMS |
| Zhun Wei, Xudong Chen, National University of Singapore, Singapore |
| |
| 13:40 - 14:00 |
| MO-SP.1P.2: GEOMETRICALLY STOCHASTIC FINITE DIFFERENCE TIME DOMAIN METHOD |
| Khadijeh Masumnia-Bisheh, Tarbiat Modares University, Iran; Cynthia Furse, University of Utah, United States |
| |
| 14:00 - 14:20 |
| MO-SP.1P.3: FAST SURROGATE MODEL-ASSISTED UNCERTAINTY QUANTIFICATION VIA QUANTIZED TENSOR TRAIN DECOMPOSITIONS |
| Luis Gomez, Duke University School of Medicine, United States; Abdulkadir Yucel, Nanyang Technological University, Singapore; Weitian Sheng, Cadence Design Systems, United States; Eric Michielssen, University of Michigan, United States |
| |
| 14:20 - 14:40 |
| MO-SP.1P.4: DEEP CONVOLUTIONAL NEURAL NETWORK APPROACH FOR SOLVING NONLINEAR INVERSE SCATTERING PROBLEMS |
| Lianlin Li, Longgang Wang, Peking University, China; Daniel Ospina Acero, Fernando Teixeira, Ohio State University, United States |
| |
| 14:40 - 15:00 |
| MO-SP.1P.5: ERROR ESTIMATION AND UNCERTAINTY QUANTIFICATION BASED ON ADJOINT METHODS IN COMPUTATIONAL ELECTROMAGNETICS |
| Branislav Notaros, Jake Harmon, Cam Key, Donald Estep, Colorado State University, United States; Troy Butler, University of Colorado Denver, United States |
| |
| 15:00 - 15:20 Break |
| |
| 15:20 - 15:40 |
| MO-SP.1P.6: A MULTI-LEVEL RECONSTRUCTION ALGORITHM FOR ELECTRICAL CAPACITANCE TOMOGRAPHY BASED ON MODULAR DEEP NEURAL NETWORKS |
| Elizabeth Chen, Costas Sarris, University of Toronto, Canada |
| |
| 15:40 - 16:00 |
| MO-SP.1P.7: DEEP NEURAL NETWORK REPRESENTATIONS OF TRANSIENT ELECTRODYNAMIC PHENOMENA |
| Oameed Noakoasteen, Shu Wang, Zhen Peng, University of New Mexico, United States |
| |
| 16:00 - 16:20 |
| MO-SP.1P.8: FAST AND ACCURATE NEAR-FIELD TO FAR-FIELD TRANSFORMATION USING AN ADAPTIVE SAMPLING ALGORITHM AND MACHINE LEARNING |
| Rezvan Rafiee Alavi, Rashid Mirzavand, Pedram Mousavi, University of Alberta, Canada |
| |
| 16:20 - 16:40 |
| MO-SP.1P.9: EXPERIMENTAL MICROWAVE TARGET IDENTIFICATION USING MACHINE LEARNING |
| Clayton Kettlewell, Kyle Hetjmanek, George Scott, Waleed Al-Shaikhli, Blake Willig, Ala-Addin Nabulsi, Somen Baidya, Reza Derakhshani, Ahmed M. Hassan, University of Missouri-Kansas City, United States |
| |
| 16:40 - 17:00 |
| MO-SP.1P.10: UNCERTAINTY QUANTIFICATION OF RADIO PROPAGATION MODELS USING ARTIFICIAL NEURAL NETWORKS |
| Aristeidis Seretis, Xingqi Zhang, Costas Sarris, University of Toronto, Canada |
| |
