A5021600103- Advanced Optimization Algorithms Research
- Polynomial and algebraic computation
- Matrix Theory and Algorithms
- Advanced Differential Equations and Dynamical Systems
- Probabilistic and Robust Engineering Design
- Forest Ecology and Biodiversity Studies
- Parallel Computing and Optimization Techniques
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Ferroelectric and Negative Capacitance Devices
- Advanced Data Storage Technologies
- Advanced Topics in Algebra
- Algebraic and Geometric Analysis
- Remote Sensing and LiDAR Applications
- Mathematical Inequalities and Applications
- Cryospheric studies and observations
- Statistical Methods and Inference
- Numerical methods in inverse problems
- Mathematics and Applications
- Mathematical functions and polynomials
- Optimization and Variational Analysis
- Commutative Algebra and Its Applications
- Advanced Statistical Methods and Models
- Arctic and Antarctic ice dynamics
- Fractional Differential Equations Solutions
University of California, San Diego
2024-2025
Beijing Institute of Big Data Research
2023-2024
Chinese Academy of Sciences
2019-2023
Academy of Mathematics and Systems Science
2019-2023
Aerospace Information Research Institute
2023
Institute of Computing Technology
2019
Snow profiles can provide reliable and detailed snow parameters for global change research, but it is still challenging to reconstruct large-scale spatiotemporally continuous profiles. Synthetic aperture radar (SAR) tomography (TomoSAR) has been proven be a promising method reconstructing the 3D structure of targets used from ground-based SAR. However, experimental condition SAR ideal algorithms developed have some problems when applied airborne spaceborne sensors, such as large...
OpenCL offers code portability but no performance portability. Given an program X specifically written for one platform P, existing compilers, which usually optimize its host and kernel codes individually, often yield poor another Q. Instead of obtaining a performance-improved version Q via manual tuning, we aim to achieve this automatically by source-to-source compiler framework, PPOpenCL. By fusing X's thread (with the operations in different work-items same work-group represented...
We consider the linear conic optimization problem with cone of nonnegative polynomials. Its dual is generalized moment problem. Moment-SOS relaxations are powerful for solving them. This paper studies finite convergence hierarchy when constraining set defined by equations whose ideal may not be real radical. Under archimedeanness, we show that has if some classical optimality conditions hold at every minimizer optimal polynomial When archimedeanness fails (this case unbounded sets), propose...
Functional reproducing kernel Hilbert spaces (FRKHSs) are appropriate function in which we seek a target from finite number of non-point-evaluation functional data. We consider reconstructing generalized moment data via regularization an FRKHS with respect to the functionals. construct specific FRKHSs and their associated kernels two classes functionals, Hamburger functionals trigonometric solve problem resulting by representer theorem. Numerical examples presented illustrate better...
Abstract This article considers the problem of We first concentrate on case that xi , i = 1, ⋯, n are binary random variable with identical probability distribution and derive a lower bound estimation. And then we observe x is normal vector obtain some estimations.
In this note, we prove that for homogeneous polynomial optimization on the sphere, if objective $f$ is generic in input space, all feasible points satisfying first order and second necessary optimality conditions are local minimizers, which addresses an issue raised recent work by Lasserre (Optimization Letters, 2021). As a corollary, implies Lasserre's hierarchy has finite convergence when generic.