A5017479931- Polynomial and algebraic computation
- Advanced Numerical Analysis Techniques
- Coding theory and cryptography
- Manufacturing Process and Optimization
- Robotic Mechanisms and Dynamics
- Computational Geometry and Mesh Generation
- Algebraic Geometry and Number Theory
- Numerical methods for differential equations
- Commutative Algebra and Its Applications
- Cryptography and Residue Arithmetic
- Advanced Differential Equations and Dynamical Systems
- Mathematics and Applications
- Advanced machining processes and optimization
- Numerical Methods and Algorithms
- Adversarial Robustness in Machine Learning
- Quantum Computing Algorithms and Architecture
- Computer Graphics and Visualization Techniques
- Model-Driven Software Engineering Techniques
- Formal Methods in Verification
- Nonlinear Waves and Solitons
- Cancer Treatment and Pharmacology
- Logic, programming, and type systems
- Constraint Satisfaction and Optimization
- 3D Shape Modeling and Analysis
- Complexity and Algorithms in Graphs
Chinese Academy of Sciences
2016-2025
Beihang University
2022-2025
Academy of Mathematics and Systems Science
2016-2025
Ningbo University of Technology
2025
University of Chinese Academy of Sciences
2017-2025
Chinese Academy of Agricultural Mechanization Sciences
2004-2024
Monash University
2022
Materials Science & Engineering
2022
First Affiliated Hospital of Jinan University
2020
Academia Sinica
1991-2008
In this paper, we propose a general scheme to analyze the gradient vanishing phenomenon, also known as barren plateau in training quantum neural networks with ZX-calculus. More precisely, extend plateaus theorem from unitary 2-design circuits any parameterized under certain reasonable assumptions. The main technical contribution of paper is representing integrations ZX-diagrams and computing them method used four concrete different structures. It shown that, for hardware efficient ansatz...
Abstract In this paper, a quantum extension of classical deep neural network (DNN) is introduced, which called QDNN and consists structured layers. It proved that the can uniformly approximate any continuous function has more representation power than DNN. Moreover, still keeps advantages DNN such as non-linear activation, multi-layer structure, efficient backpropagation training algorithm. Furthermore, uses parameterized circuits (PQCs) basic building blocks hence be used on near-term noisy...
The multi-phase redundant permanent magnet machine is essentially a synchronous motor (PMSM) with multiple sets of windings arranged on the stator. This configuration facilitates fault-tolerant operation PMSM in event one or phase failing. In response to practical requirements electromechanical actuation systems aviation industry, this paper proposed solution for high-power, high-speed machine. stator and rotor machine, along short-circuit current suppression, are analysed, leading design...
Article Free Access Share on Solving parametric algebraic systems Authors: Xiao-Shan Gao View Profile , Shang-Ching Chou Authors Info & Claims ISSAC '92: Papers from the international symposium Symbolic and computationAugust 1992 Pages 335–341https://doi.org/10.1145/143242.143348Published:01 August 1992Publication History 30citation303DownloadsMetricsTotal Citations30Total Downloads303Last 12 Months16Last 6 weeks3 Get Citation AlertsNew Alert added!This alert has been successfully added will...
With the use of computer algebra, method that straightforwardly leads to travelling wave solutions is presented. The compound KdV–Burgers equation and KP–B are chosen illustrate this approach. As a result, their abundant new soliton-like period form found.
We give a necessary and sufficient condition for an algebraic ODE to have rational type general solution. For autonomous first order ODE, we algorithm compute solution if it exists. The is based on the relation between solutions of parametrizations plane curve defined by Padé approximants.