A5087963011- Probabilistic and Robust Engineering Design
- Gaussian Processes and Bayesian Inference
- Advanced Multi-Objective Optimization Algorithms
- Quantum Computing Algorithms and Architecture
- Model Reduction and Neural Networks
- Sparse and Compressive Sensing Techniques
- Quantum Information and Cryptography
- Fractional Differential Equations Solutions
- Spectroscopy and Quantum Chemical Studies
- Structural Health Monitoring Techniques
- Optimal Power Flow Distribution
- Advanced Battery Technologies Research
- Nanofluid Flow and Heat Transfer
- Machine Learning in Materials Science
- Image and Signal Denoising Methods
- Numerical methods in engineering
- Energy Load and Power Forecasting
- Advancements in Semiconductor Devices and Circuit Design
- Seismic Imaging and Inversion Techniques
- Advanced Fluorescence Microscopy Techniques
- Differential Equations and Numerical Methods
- Power System Optimization and Stability
- Fluid Dynamics and Vibration Analysis
- Music and Audio Processing
- Stochastic Gradient Optimization Techniques
Lehigh University
2018-2025
Shanghai University of Electric Power
2021-2024
Shandong University
2017-2024
Fujian Medical University
2024
University of Utah
2022
Chinese University of Hong Kong, Shenzhen
2022
The University of Texas at Dallas
2022
University of Kentucky
2022
Michigan State University
2022
Pacific Northwest National Laboratory
2014-2020
Hierarchical uncertainty quantification can reduce the computational cost of stochastic circuit simulation by employing spectral methods at different levels. This paper presents an efficient framework to simulate hierarchically some challenging circuits/systems that include high-dimensional subsystems. Due high parameter dimensionality, it is both extract surrogate models low level design hierarchy and handle them in high-level simulation. In this paper, we develop ANOVA-based circuit/MEMS...
In the study of subsurface seismic imaging, solving acoustic wave equation is a pivotal component in existing models. The advancement deep learning enables partial differential equations, including by applying neural networks to identify mapping between inputs and solution. This approach can be faster than traditional numerical methods when numerous instances are solved. Previous works that concentrate on consider either single velocity model or multiple simple models, which restricted...
We construct effective coarse-grained (CG) models for polymeric fluids by employing two coarse-graining strategies. The first one is a forward-coarse-graining procedure the Mori-Zwanzig (MZ) projection while other applies reverse-coarse-graining procedure, such as iterative Boltzmann inversion (IBI) and stochastic parametric optimization (SPO). More specifically, we perform molecular dynamics (MD) simulations of star polymer melts to provide atomistic fields be coarse-grained. Each molecule...
Mechanical instability has been shown to play an important role in the formation of wrinkle structures biofilms, which not only can adopt modes as templates regulate their 3D architectures but also tune internal stresses achieve stable patterns. Inspired by nature, we report a mechanical–chemical coupling method fabricate free-standing conducting films with instability-driven hierarchical micro/nanostructured When polypyrrole (PPy) film is grown on elastic substrate via chemical oxidation...
This study introduces a novel Gaussian process (GP) regression framework that probabilistically enforces physical constraints, with particular focus on equality conditions. The GP model is trained using the quantum-inspired Hamiltonian Monte Carlo (QHMC) algorithm, which efficiently samples from wide range of distributions by allowing particle's mass matrix to vary according probability distribution. By integrating QHMC into probabilistic handling approach balances computational cost and...
This study introduces a novel Gaussian process (GP) regression framework that probabilistically enforces physical constraints, with particular focus on equality conditions. The GP model is trained using the quantum-inspired Hamiltonian Monte Carlo (QHMC) algorithm, which efficiently samples from wide range of distributions by allowing particle’s mass matrix to vary according probability distribution. By integrating QHMC into probabilistic handling this approach balances computational cost...
In this work, we propose a new Gaussian process (GP) regression framework that enforces the physical constraints in probabilistic manner. Specifically, focus on inequality and monotonicity constraints. This GP model is trained by quantum-inspired Hamiltonian Monte Carlo (QHMC) algorithm, which an efficient way to sample from broad class of distributions allowing particle have random mass matrix with probability distribution. Integrating QHMC into constrained sense, our approach enhances...
As quantum computers evolve, simulations of programs on classical will be essential in validating algorithms, understanding the effect system noise, and designing applications for future computers. In this paper, we first propose a new multi-GPU programming methodology called MG-BSP which constructs virtual BSP machine top modern platforms, apply to build density matrix simulator DM-Sim. We formulation that can significantly reduce communication overhead, show formula transformation conserve...
Sensor placement at the extrema of empirical orthogonal functions (EOFs) is efficient and leads to accurate reconstruction ocean state from a limited number measurements. In this paper, we develop important new extensions approach that optimize sensor avoid redundant measurements, employ imperfect EOF modes, take into account measurement errors. We use simulation outputs Finite Volume Community Ocean Model applied Nantucket Sound region evaluate performances compare it against other similar...
Process variations are a major concern in today's chip design since they can significantly degrade performance. To predict such degradation, existing circuit and MEMS simulators rely on Monte Carlo algorithms, which typically too slow. Therefore, novel fast stochastic highly desired. This paper first reviews our recently developed testing simulator that achieve speedup factors of hundreds to thousands over Carlo. Then, we develop hierarchical spectral simulate complex or system consisting...
Gaussian process (GP) regression is a flexible non-parametric approach to approximate complex models. In many cases, these models correspond processes with bounded physical properties. Standard GP typically results in proxy model which unbounded for all temporal or spacial points, and thus leaves the possibility of taking on infeasible values. We propose an enforce constraints probabilistic way under framework. addition, this new reduces variance resulting model.
Quantum linear system algorithms (QLSAs) have the potential to speed up that rely on solving systems. Interior point methods (IPMs) yield a fundamental family of polynomial-time for optimization problems. IPMs solve Newton at each iteration compute search direction; thus, QLSAs can potentially IPMs. Due noise in contemporary quantum computers, quantum-assisted (QIPMs) only admit an inexact solution system. Typically, direction leads infeasible solution, so, overcome this, we propose...
Biomolecules exhibit conformational fluctuations near equilibrium states, inducing uncertainty in various biological properties a dynamic way. We have developed general method to quantify the of target induced by fluctuations. Using generalized polynomial chaos (gPC) expansion, we construct surrogate model property with respect varying states. also propose increase sparsity gPC expansion defining set "active space" random variables. With increased sparsity, employ compressive sensing...
We develop a new mathematical framework to study the optimal design of air electrode microstructures for lithium-oxygen (Li-O2) batteries. The parameters characterize an air-electrode microstructure include porosity, surface-to-volume ratio, and associated with pore-size distribution. A surrogate model (also known as response surface) discharge capacity is first constructed function these parameters. In particular, Gaussian process regression method, co-kriging, employed due its accuracy...