Publications

StructMoE: Structured Mixture of Experts Using Low Rank Experts
We introduce StructMoE, a method to scale MoE architectures by augmenting experts with dynamic capacity using structured matrices we …
Zain Sarwar, Ashwinee Panda, Benjamin Thérien, Stephen Rawls, Anirban Das, Kartik Balasubramaniam, Berkcan Kapusuzoglu, Shixiong Zhang, Sambit Sahu, Milind Naphade, Supriyo Chakraborty
Posted Jan 1, 1001
Link
Multi-Level Bayesian Calibration of a Multi-Component Dynamic System Model
This paper proposes a multi-level Bayesian calibration approach that fuses information from heterogeneous sources and accounts for …
Berkcan Kapusuzoglu, Sankaran Mahadevan, Shunsaku Matsumoto, Yoshitomo Miyagi, Daigo Watanabe
Posted Jan 1, 1001
Link
Adaptive surrogate modeling for high-dimensional spatio-temporal output
This paper develops an adaptive surrogate modeling method for problems with very high-dimensional spatio-temporal outputs. The analysis …
Berkcan Kapusuzoglu, Sankaran Mahadevan, Shunsaku Matsumoto, Yoshitomo Miyagi, Daigo Watanabe
Posted Jan 1, 1001
Link
Dimension reduction for efficient surrogate modeling in high-dimensional applications
View Video Presentation: https://doi.org/10.2514/6.2022-1440.vidEngineering analysis using finite element models of complex aerospace …
Berkcan Kapusuzoglu, Yulin Guo, Sankaran Mahadevan, Shunsaku Matsumoto, Miyagi Yoshitomo, Shunsuke Taba, Daigo Watanabe
Posted Jan 1, 1001
Link
Multi-objective optimization under uncertainty of part quality in fused filament fabrication
This work presents a data-driven methodology for multi-objective optimization under uncertainty of process parameters in the fused …
Berkcan Kapusuzoglu, Paromita Nath, Matthew Sato, Sankaran Mahadevan, Paul Witherell
Posted Jan 1, 1001
Link
Information fusion and machine learning for sensitivity analysis using physics knowledge and experimental data
When computational models (either physics-based or data-driven) are used for the sensitivity analysis of engineering systems, the …
Berkcan Kapusuzoglu, Sankaran Mahadevan
Posted Jan 1, 1001
Link
Process Optimization under Uncertainty for Improving the Bond Quality of Polymer Filaments in Fused Filament Fabrication
This paper develops a computational framework to optimize the process parameters such that the bond quality between extruded polymer …
Paul Witherell, Berkcan Kapusuzoglu, Matthew Sato, Sankaran Mahadevan
Posted Jan 1, 1001
Link
Digital twin approach for damage-tolerant mission planning under uncertainty
The digital twin paradigm that integrates the information obtained from sensor data, physics models, as well as operational and …
Pranav M Karve, Yulin Guo, Berkcan Kapusuzoglu, Sankaran Mahadevan, Mulugeta a Haile
Posted Jan 1, 1001
Link
Digital twin approach for intelligent operation planning and health management of mechanical systems
The digital twin paradigm aims to fuse information obtained from sensor data, physics models, and operational data for a mechanical …
Pranav Karve, Yulin Guo, Berkcan Kapusuzoglu, Sankaran Mahadevan, Mulugeta Haile
Posted Jan 1, 1001
Link
Discrete eigenseparation-based reduced order homogenization method for failure modeling of composite materials
This manuscript presents a new reduced order computational homogenization method for failure analysis of composite materials. The …
Caglar Oskay, Zimu Su, Berkcan Kapusuzoglu
Posted Jan 1, 1001
Link