Manifold differential evolution (MDE)

LiuYong-Jin,XuChun-Xu,YiRan,FanDian,HeYing

IF: 6.2

2016-01-01

ACM Transactions on Graphics

Abstract:Computing centroidal Voronoi tessellations (CVT) has many applications in computer graphics. The existing methods, such as the Lloyd algorithm and the quasi-Newton solver, are efficient and easy to...

What problem does this paper attempt to address?

Manifold Differential Evolution (MDE)

Yong-Jin Liu,Xu,Ran Yi,Dian Fan,Ying He

DOI: https://doi.org/10.1145/2980179.2982424

IF: 6.2

2016-01-01

ACM Transactions on Graphics

Abstract:Computing centroidal Voronoi tessellations (CVT) has many applications in computer graphics. The existing methods, such as the Lloyd algorithm and the quasi-Newton solver, are efficient and easy to implement; however, they compute only the local optimal solutions due to the highly non-linear nature of the CVT energy. This paper presents a novel method, called manifold differential evolution (MDE), for computing globally optimal geodesic CVT energy on triangle meshes. Formulating the mutation operator using discrete geodesics, MDE naturally extends the powerful differential evolution framework from Euclidean spaces to manifold domains. Under mild assumptions, we show that MDE has a provable probabilistic convergence to the global optimum. Experiments on a wide range of 3D models show that MDE consistently out-performs the existing methods by producing results with lower energy. Thanks to its intrinsic and global nature, MDE is insensitive to initialization and mesh tessellation. Moreover, it is able to handle multiply-connected Voronoi cells, which are challenging to the existing geodesic CVT methods.
Differential evolution with M-fitness method

Ying Yang,Min Yao

DOI: https://doi.org/10.1109/ICCCNT.2014.6963045

2014-01-01

Abstract:Differential Evolution(DE) is a powerful algorithm to solve global optimization problems. Because the optimization process of original DE is quite easy to understand and code, it has been widely applied in many fields. In recent years, many adaptive parameters DEs have been proposed and achieved better performance on many problems. But simplicity and parallelism of DE have been decreased in those adaptive DE, so they can't be easily transferred to other fields. Moreover, adaptive parameter mechanisms don't always perform better compared with some popular parameter settings. To enhance the performance while maintaining the simplicity and parallelism of DE algorithm, in this paper, we introduce a m-fitness method. The method we proposed use distribution information of fitness value to tune p-value which is a parameter used in DE/pbest/1 to control convergence speed. Moreover, in the method, the information also has been used in selection phase by using half-meanfit selection we proposed in paper. DE with m-fitness method(mDE) is compared on benchmark functions with classical DE and some representative adaptive DE. The results show that the DE with m-fitness method is competitive with other various DE in performance.
Performance Study of Differential Evolution with Various Mutation Strategies Applied to Water Distribution System Optimization

Feifei Zheng,Angus R. Simpson,Aaron Zecchin

DOI: https://doi.org/10.1061/41173(414)18

2011-01-01

Abstract:Differential evolution (DE) is a relatively new optimization technique that has been employed to optimize the design of water distribution systems (WDSs). There are three important operators involved in the use of DE: mutation, crossover and selection. These operators are similar to the commonly used genetic algorithms (GAs). However, DE differs significantly from GAs in that mutation is an important operator for DE, while in contrast, crossover is an important operator for GAs. It has been found that the success of the DE algorithm in solving different mathematical optimization problems crucially depends on the mutation strategy that is used. This paper aims to investigate the relative effectiveness of five frequently used mutation strategies of DE when applied to WDS optimization. The five DE variants with different mutation strategies are applied to two well-known WDS case studies: the New York Tunnels Problem and the Hanoi Problem.
MASSIVE: A Multidisciplinary Applications-Oriented Simulation and Visualization Environment

Guiyi Wei,Guanghua Song,Yao Zheng,Cuiju Luan,Chaoyan Zhu,Wei Wang

DOI: https://doi.org/10.1109/SCC.2004.1358069

2004-01-01

Abstract:The objective of the MASSIVE project is to build a Grid platform to enable large-scale and distributed scientific and engineering computations. Its focus is numerical simulation and visualization. This paper addresses a prototype Grid utilizing resources at the Center for Engineering and Scientific Computation (CESC) in Zhejiang University (ZJU). This Grid is capable of taking a model geometry generated by CAD systems, transferring it to an SGI Onyx 3900 supercomputer where it is meshed. The meshed geometry is then transferred to a Dawning PC cluster where it is used to solve computational fluid dynamics and computational solid mechanics problems and it is finally visualized showing numerical results.
Mesh Generation for FEM Based on Centroidal Voronoi Tessellations

Yu-xin Jie,Xu-dong Fu,Yan Liu

DOI: https://doi.org/10.1016/j.matcom.2013.05.014

IF: 3.601

2014-01-01

Mathematics and Computers in Simulation

Abstract:Centroidal Voronoi tessellations (CVTs) are very useful in a variety of applications, which can be used in triangular or tetrahedral mesh generations. There are several algorithms for determining CVTs, including MacQueen's method, Lloyd's method, and generalized probabilistic Lloyd's method. The latter is a combination of MacQueen's method and Lloyd's method, which is thought to be one of the most efficient methods to determine high-quality CVTs without the need to explicitly construct Voronoi diagrams. However, the convergence of these methods is difficult to achieve, since they are inclined to be trapped at local minima of cost functional. In this paper, simulated annealing (SA) is introduced to overcome this problem, which is applied to make mesh generation in domains including convex domains, a concaved domain, a multi-connected domain, and a circular domain. The efficiency of this method, and 2-D and 3-D mesh generations are successfully verified through examples.
Delaunay Mesh Simplification with Differential Evolution.

Ran Yi,Yong-Jin Liu,Ying He

DOI: https://doi.org/10.1145/3272127.3275068

IF: 6.2

2018-01-01

ACM Transactions on Graphics

Abstract:Delaunay meshes (DM) are a special type of manifold triangle meshes --- where the local Delaunay condition holds everywhere --- and find important applications in digital geometry processing. This paper addresses the general DM simplification problem: given an arbitrary manifold triangle mesh M with n vertices and the user-specified resolution m (< n), compute a Delaunay mesh M* with m vertices that has the least Hausdorffdistance to M. To solve the problem, we abstract the simplification process using a 2D Cartesian grid model, in which each grid point corresponds to triangle meshes with a certain number of vertices and a simplification process is a monotonic path on the grid. We develop a novel differential-evolution-based method to compute a low-cost path, which leads to a high quality Delaunay mesh. Extensive evaluation shows that our method consistently outperforms the existing methods in terms of approximation error. In particular, our method is highly effective for small-scale CAD models and man-made objects with sharp features but less details. Moreover, our method is fully automatic and can preserve sharp features well and deal with models with multiple components, whereas the existing methods often fail.
Geodesic Centroidal Voronoi Tessellations: Theories, Algorithms and Applications

Zipeng Ye,Ran Yi,Minjing Yu,Yong-Jin Liu,Ying He

DOI: https://doi.org/10.48550/arXiv.1907.00523

2019-07-01

Graphics

Abstract:Nowadays, big data of digital media (including images, videos and 3D graphical models) are frequently modeled as low-dimensional manifold meshes embedded in a high-dimensional feature space. In this paper, we summarized our recent work on geodesic centroidal Voronoi tessellations(GCVTs), which are intrinsic geometric structures on manifold meshes. We show that GCVT can find a widely range of interesting applications in computer vision and graphics, due to the efficiency of search, location and indexing inherent in these intrinsic geometric structures. Then we present the challenging issues of how to build the combinatorial structures of GCVTs and establish their time and space complexities, including both theoretical and algorithmic results.
Digital Differential Geometry Processing

Xin-Guo Liu,Hu-Jun Bao,Qun-Sheng Peng

DOI: https://doi.org/10.1007/s11390-006-0847-5

IF: 1.871

2006-01-01

Journal of Computer Science and Technology

Abstract:The theory and methods of digital geometry processing has been a hot research area in computer graphics, as geometric models serves as the core data for 3D graphics applications. The purpose of this paper is to introduce some recent advances in digital geometry processing, particularly mesh fairing, surface parameterization and mesh editing, that heavily use differential geometry quantities. Some related concepts from differential geometry, such as normal, curvature, gradient, Laplacian and their counterparts on digital geometry are also reviewed for understanding the strength and weakness of various digital geometry processing methods.
Manifold Diffusion Fields

Ahmed A. Elhag,Yuyang Wang,Joshua M. Susskind,Miguel Angel Bautista

2024-01-20

Abstract:We present Manifold Diffusion Fields (MDF), an approach that unlocks learning of diffusion models of data in general non-Euclidean geometries. Leveraging insights from spectral geometry analysis, we define an intrinsic coordinate system on the manifold via the eigen-functions of the Laplace-Beltrami Operator. MDF represents functions using an explicit parametrization formed by a set of multiple input-output pairs. Our approach allows to sample continuous functions on manifolds and is invariant with respect to rigid and isometric transformations of the manifold. In addition, we show that MDF generalizes to the case where the training set contains functions on different manifolds. Empirical results on multiple datasets and manifolds including challenging scientific problems like weather prediction or molecular conformation show that MDF can capture distributions of such functions with better diversity and fidelity than previous approaches.

Machine Learning
Computing Variation Modes for Point Set Surfaces.

Lanfang Miao,Jin Huang,Xinguo Liu,Hujun Bao,Qunsheng Peng,Baining Guo

DOI: https://doi.org/10.1109/pbg.2005.194065

2005-01-01

Abstract:Point sets have become a popular shape representation. In this paper, we present a novel approach to computing variation modes for point set surfaces, and represent the point set surface as a linear combination of the variation modes, called a generative representation for the point set surface. Given a point set, our approach consists of two steps: the first is to produce a set of new samples with increasing smoothness and less detailed features. We use a modified smoothing method based on moving least squares (MLS) surface to produce the samples. The second is to arrange the shape vectors of the new samples together with the original point set into a matrix, and then compute the singular value decomposition of the matrix, producing a set of variation modes (the eigen vectors). Using the variation modes and the generative representation, we can easily synthesize new shapes. Typical applications are low/high/band pass filtering as well as denoising and detail enhancement in multiple scales.
Flattenable Mesh Processing by Controllable Laplacian Evolution

Hongwei Lin,Yunbo Zhang,Charlie C. L. Wang,Shuming Gao

DOI: https://doi.org/10.1115/detc2010-28301

2010-01-01

Abstract:Models represented by polygonal meshes have been more and more widely used in CAD/CAM systems. In sheet manufacturing industries, the flattenability of a model is very important. Prior methods for processing the flattenability of a mesh surface usually employ a constrained optimization framework, which takes the positions of all its non-boundary vertices as variables in computation. For a mesh surface with hundred thousands of vertices, solving such an optimization is very time-consuming, and may exceed the capacity of main memory. In this paper, we develop a controllable evolution method to process the flattenability of a given mesh patch. It decouples the global optimization problem into a sequence of local controllable evolution steps, each of which has only one variable. Therefore, mesh surfaces with large number of vertices can be processed.
Study of calculating the dynamic available transfer capability with the differential evolution algorithm

WANG Jun,CAI Xing-guo

2010-01-01

Abstract:The mathematic model of dynamic available transfer capability is established based on the optimal power flow considering the transient stability constraints. The differential evolution algorithm (DE) is proposed which is used in solving available transfer capability (ATC) calculation considering the transient stability constraints. A composite fitness function method, which does not require any penalty coefficient, is employed for constraint handling. The DE has such features such as simple to use, little controlled parameters and carry out random, parallel and direct global search, which is easy to understand and put into effect, and also it has good convergence property and fast calculating speed. The DE is an effective and robustness method in solving the nonlinear, non-differentiable, multi-peaks, higher dimensional and complex functions. The availability and rationality of DE is verified by the calculation results of the WSCC9 bus system.
Developments and Design of Differential Evolution Algorithm for Non-linear/Non-convex Engineering Optimization

Tiwari, Pooja,Mishra, Vishnu Narayan,Parouha, Raghav Prasad

DOI: https://doi.org/10.1007/s11831-023-10036-9

IF: 9.7

2024-01-11

Archives of Computational Methods in Engineering

Abstract:Nowadays, the differential evolution (DE) achieved noticeable progress and solved a wide range of non-linear/non-convex engineering optimization issues. As a strong optimizer, DE has many advantages like simple structure, strong exploitation ability and considerable convergence speed. However, DE also suffers from low diversification, poor exploration ability and stagnation. After significant reviewing and to avoid aforesaid gaps, this paper reports a modified DE (mDE) for solving non-linear/non-convex optimization problems, especially non-convex economic dispatch (ED). It adopted novel mutation strategy with new control parameters using concept of particle swarm optimization (PSO), to enhance exploration and exploitation activities more profusely and increase the global search capability. Also, a new crossover rate is employed in mDE, to attain higher convergence accuracy and quality optimal solutions. Finally, a novel selection strategy is introduced in mDE, to facilitate information sharing as well as for escaping local minima and keeps evolving. To validate the mDE performance, IEEE CEC2006 non-linear constrained benchmark suites are solved. Furthermore, its practicality, efficacy and excellence are further demonstrated in six complex non-linear engineering optimization issues and six cases (3, 6, 13, 15, 40 and 140—unit system) of non-convex ED problems. Besides, the mDE results are compared with other peer meta-heuristic methods. Comparison results analysis validates the competitive presentation of the developed mDE.

computer science, interdisciplinary applications,engineering, multidisciplinary,mathematics
Differential evolution for population diversity mechanism based on covariance matrix

Xueying Shao,Yihong Ding

DOI: https://doi.org/10.1016/j.isatra.2023.06.023

Abstract:Differential evolution (DE) is a heuristic global search algorithm based on population. It has exhibited great adaptability in solving continuous-domain problems, but sometimes suffered from insufficient local search ability and being trapped in local optimum when dealing with complicated optimization problems. To solve these problems, an improved differential evolution algorithm with population diversity mechanism based on covariance matrix (CM-DE) is proposed. First, a new parameter adaptation strategy is used to adapt the control parameters, in which the scale factor F is updated according to the improved wavelet basis function in the early stage and Cauchy distribution in the later stage and the crossover rate CR is generated according to normal distribution. The diversity of population and convergence speed are improved by employing the method above. Second, the perturbation strategy is incorporated into crossover operator to enhance the search ability of DE. Finally, the covariance matrix of the population is constructed, where the variance in the covariance matrix is used as indicator to measure the similarity between individuals in the population in order to prevent the algorithm from falling into local optimum resulted by low population diversity. The CM-DE is compared with the state-of-art DE variants including LSHADE (Tanabe and Fukunaga, 2014), jSO [1], LPalmDE [2], PaDE [3] and LSHADE-cnEpSin [4] under 88 test functions from CEC2013 [5], CEC2014 [6] and CEC2017 (Wu et al., 2017) test suites. From the experiment results, it is obvious that among 30 benchmark functions from CEC2017 on 50D optimization, the CM-DE algorithm has 22, 20, 24, 23, 28 better performances comparing with LSHADE, jSO, LPalmDE, PaDE, and LSHADE-cnEpsin. For CEC2017 on 30D optimization, the proposed algorithm secures better performance on 19 out of 30 benchmark functions in terms of convergence speed. In addition, a real-world application is also used to verify the feasibility of the proposed algorithm. The experiment results validate the highly competitive performance in terms of solution accuracy and convergence speed.
MEG3D——An Open-Source Software for DFN Model and 3D Numerical Manifold Elements Generation

Defu Tong,Xiongwei Yi,Fei Tan,Yuyong Jiao

DOI: https://doi.org/10.1016/j.compgeo.2024.106383

IF: 5.218

2024-05-16

Computers and Geotechnics

Abstract:Numerical manifold method (NMM) has shown its ability to solve continuous and discontinuous deformation problems in a unified framework. However, due to the complexity of geometry description and the absence of a reliable 3D contact algorithm, the development of 3D-NMM still has a long-time challenge. In this study, an open-source software named MEG3D to generate the numerical model for 3D-NMM is developed. The MEG3D is a fast, light, and user-friendly interactive software to identify geometry, generate joint networks, generate structured finite mathematical mesh, cut blocks, and generate numerical manifold elements (MEs). In this software, a new C++ programming strategy with high modularization and good portability, and a novel data storage format and data structure were used. Based on the OpenGL library, real-time visualization and interactive interfaces were built. Examples containing curve block cutting, discrete fracture networks (DFN) model, soil-rock mixture slopes, complex shaped geometry, etc. have indicated that the software is robust, efficient, and user-friendly. This software can also generate models for the discrete element method (DEM), discontinuous deformation analysis (DDA), etc. Therefore, the MEG3D can be used as a general pre-processing program for 3D-NMM and other block-based numerical methods.

computer science, interdisciplinary applications,geosciences, multidisciplinary,engineering, geological
Gradient of the Objective Function for an Anisotropic Centroidal Voronoi Tessellation (CVT) - A revised, detailed derivation

Giacomo Parigi,Marco Piastra

DOI: https://doi.org/10.48550/arXiv.1408.5622

2014-08-25

Abstract:In their recent article (2010), Levy and Liu introduced a generalization of Centroidal Voronoi Tessellation (CVT) - namely the Lp-CVT - that allows the computation of an anisotropic CVT over a sound mathematical framework. In this article a new objective function is defined, and both this function and its gradient are derived in closed-form for surfaces and volumes. This method opens a wide range of possibilities, also described in the paper, such as quad-dominant surface remeshing, hex-dominant volume meshing or fully-automated capturing of sharp features. However, in the same paper, the derivations of the gradient and of the new objective function are only partially expanded in the appendices, and some relevant requisites on the anisotropy field are left implicit. In order to better harness the possibilities described there, in this work the entire derivation process is made explicit. In the authors' opinion, this also helps understanding the working conditions of the method and its possible applications.

Computational Geometry
DE-Path: A Differential-Evolution-Based Method for Computing Energy-Minimizing Paths on Surfaces

Zipeng Ye,Yong-Jin Liu,Jianmin Zheng,Kai Hormann,Ying He

DOI: https://doi.org/10.1016/j.cad.2019.05.025

IF: 3.652

2019-01-01

Computer-Aided Design

Abstract:Computing energy-minimizing paths that are general for different energy forms is a common task in science and engineering. Conventional methods adopt numerical solvers, such as conjugate gradient or quasi-Newton. While these are efficient, the results are highly sensitive with respect to the initial paths. In this paper we develop a method based on differential evolution (DE) for computing optimal solutions. We propose a simple strategy to encode paths and define path operations, such as addition and scalar multiplication, so that the discrete paths can fit into the DE framework. We demonstrate the effectiveness of our method on three applications: (1) computing discrete geodesic paths on surfaces with non-uniform density function; (2) finding a smooth path that follows a given vector field as much as possible; and (3) finding a curve on a terrain with (near-) constant slope.
Covariance and crossover matrix guided differential evolution for global numerical optimization

YongLi Li,JinFu Feng,JunHua Hu

DOI: https://doi.org/10.1186/s40064-016-2838-5

2016-07-26

SpringerPlus

Abstract:Differential evolution (DE) is an efficient and robust evolutionary algorithm and has wide application in various science and engineering fields. DE is sensitive to the selection of mutation and crossover strategies and their associated control parameters. However, the structure and implementation of DEs are becoming more complex because of the diverse mutation and crossover strategies that use distinct parameter settings during the different stages of the evolution. A novel strategy is used in this study to improve the crossover and mutation operations. The crossover matrix, instead of a crossover operator and its control parameter CR, is proposed to implement the function of the crossover operation. Meanwhile, Gaussian distribution centers the best individuals found in each generation based on the proposed covariance matrix, which is generated between the best individual and several better individuals. Improved mutation operator based on the crossover matrix is randomly selected to generate the trial population. This operator is used to generate high-quality solutions to improve the capability of exploitation and enhance the preference of exploration. In addition, the memory population is randomly chosen from previous generation and used to control the search direction in the novel mutation strategy. Accordingly, the diversity of the population is improved. Thus, CCDE, which is a novel efficient and simple DE variant, is presented in this paper. CCDE has been tested on 30 benchmarks and 5 real-world optimization problems from the IEEE Congress on Evolutionary Computation (CEC) 2014 and CEC 2011, respectively. Experimental and statistical results demonstrate the effectiveness of CCDE for global numerical and engineering optimization. CCDE can solve the test benchmark functions and engineering problems more successfully than the other DE variants and algorithms from CEC 2014.
Efficient mini-batch stochastic gradient descent with Centroidal Voronoi Tessellation for PDE-constrained optimization under uncertainty

Liuhong Chen,Meixin Xiong,Ju Ming,Xiaoming He

DOI: https://doi.org/10.1016/j.physd.2024.134216

IF: 3.751

2024-05-26

Physica D Nonlinear Phenomena

Abstract:The study of optimal control problems under uncertainty plays an important role in scientific numerical simulations. This class of optimization problems is frequently utilized in engineering, biology and finance. Although stochastic gradient descent is a well-known stochastic optimization technique for solving the approximate optimal control problem, it often exhibits a slow convergence rate due to the inherent variance in gradient approximation. To address this issue, we propose a more efficient stochastic optimization algorithm based on Centroidal Voronoi Tessellation sampling. This strategy reduces the error in gradient estimation compared to the conventional mini-batch stochastic gradient descent method. Our approach involves dividing the whole snapshot set into several Voronoi cells with low variance and extracting samples with good uniformity from each region to construct an unbiased estimation of the full gradient. Our method can economically and stably approximate numerical optimal control functions even with a fixed step size. Numerical results demonstrate that the proposed method is a reliable algorithm with great potential for applications in the field of optimization.

mathematics, applied,physics, mathematical, multidisciplinary, fluids & plasmas
Solving Partial Differential Equations Numerically on Manifolds with Arbitrary Spatial Topologies

Lee Lindblom,Bela Szilagyi

DOI: https://doi.org/10.1016/j.jcp.2013.02.031

2012-10-18

Abstract:A multi-cube method is developed for solving systems of elliptic and hyperbolic partial differential equations numerically on manifolds with arbitrary spatial topologies. It is shown that any three-dimensional manifold can be represented as a set of non-overlapping cubic regions, plus a set of maps to identify the faces of adjoining regions. The differential structure on these manifolds is fixed by specifying a smooth reference metric tensor. Matching conditions that ensure the appropriate levels of continuity and differentiability across region boundaries are developed for arbitrary tensor fields. Standard numerical methods are then used to solve the equations with the appropriate boundary conditions, which are determined from these inter-region matching conditions. Numerical examples are presented which use pseudo-spectral methods to solve simple elliptic equations on multi-cube representations of manifolds with the topologies T^3, S^2 x S^1 and S^3. Examples are also presented of numerical solutions of simple hyperbolic equations on multi-cube manifolds with the topologies R x T^3, R x S^2 x S^1 and R x S^3.

Computational Physics

Manifold differential evolution (MDE)

Manifold Differential Evolution (MDE)

Differential evolution with M-fitness method

Performance Study of Differential Evolution with Various Mutation Strategies Applied to Water Distribution System Optimization

MASSIVE: A Multidisciplinary Applications-Oriented Simulation and Visualization Environment

Mesh Generation for FEM Based on Centroidal Voronoi Tessellations

Delaunay Mesh Simplification with Differential Evolution.

Geodesic Centroidal Voronoi Tessellations: Theories, Algorithms and Applications

Digital Differential Geometry Processing

Manifold Diffusion Fields

Computing Variation Modes for Point Set Surfaces.

Flattenable Mesh Processing by Controllable Laplacian Evolution

Study of calculating the dynamic available transfer capability with the differential evolution algorithm

Developments and Design of Differential Evolution Algorithm for Non-linear/Non-convex Engineering Optimization

Differential evolution for population diversity mechanism based on covariance matrix

MEG3D——An Open-Source Software for DFN Model and 3D Numerical Manifold Elements Generation

Gradient of the Objective Function for an Anisotropic Centroidal Voronoi Tessellation (CVT) - A revised, detailed derivation

DE-Path: A Differential-Evolution-Based Method for Computing Energy-Minimizing Paths on Surfaces

Covariance and crossover matrix guided differential evolution for global numerical optimization

Efficient mini-batch stochastic gradient descent with Centroidal Voronoi Tessellation for PDE-constrained optimization under uncertainty

Solving Partial Differential Equations Numerically on Manifolds with Arbitrary Spatial Topologies