scholarly journals Multi-view Clustering via Late Fusion Alignment Maximization

2019 ◽  
Author(s):  
Siwei Wang ◽  
Xinwang Liu ◽  
En Zhu ◽  
Chang Tang ◽  
Jiyuan Liu ◽  
...  
Keyword(s):  
Computational Complexity ◽  
Iterative Algorithm ◽  
Optimization Problem ◽  
Optimization Procedure ◽  
Multiple Views ◽  
Late Fusion ◽  
Complementary Information ◽  
Benchmark Datasets ◽  
Effectiveness And Efficiency ◽  
Consensus Partition

Multi-view clustering (MVC) optimally integrates complementary information from different views to improve clustering performance. Although demonstrating promising performance in many applications, we observe that most of existing methods directly combine multiple views to learn an optimal similarity for clustering. These methods would cause intensive computational complexity and over-complicated optimization. In this paper, we theoretically uncover the connection between existing k-means clustering and the alignment between base partitions and consensus partition. Based on this observation, we propose a simple but effective multi-view algorithm termed {Multi-view Clustering via Late Fusion Alignment Maximization (MVC-LFA)}. In specific, MVC-LFA proposes to maximally align the consensus partition with the weighted base partitions. Such a criterion is beneficial to significantly reduce the computational complexity and simplify the optimization procedure. Furthermore, we design a three-step iterative algorithm to solve the new resultant optimization problem with theoretically guaranteed convergence. Extensive experiments on five multi-view benchmark datasets demonstrate the effectiveness and efficiency of the proposed MVC-LFA.

scholarly journals GSPL: A Succinct Kernel Model for Group-Sparse Projections Learning of Multiview Data

2021 ◽  
Author(s):  
Danyang Wu ◽  
Jin Xu ◽  
Xia Dong ◽  
Meng Liao ◽  
Rong Wang ◽  
...  
Keyword(s):  
Feature Selection ◽  
Iterative Algorithm ◽  
Optimization Problem ◽  
Selection Task ◽  
Experimental Results ◽  
Multiple Views ◽  
Intrinsic Parameters ◽  
Norm Constraint ◽  
Kernel Model

This paper explores a succinct kernel model for Group-Sparse Projections Learning (GSPL), to handle multiview feature selection task completely. Compared to previous works, our model has the following useful properties: 1) Strictness: GSPL innovatively learns group-sparse projections strictly on multiview data via ‘2;0-norm constraint, which is different with previous works that encourage group-sparse projections softly. 2) Adaptivity: In GSPL model, when the total number of selected features is given, the numbers of selected features of different views can be determined adaptively, which avoids artificial settings. Besides, GSPL can capture the differences among multiple views adaptively, which handles the inconsistent problem among different views. 3) Succinctness: Except for the intrinsic parameters of projection-based feature selection task, GSPL does not bring extra parameters, which guarantees the applicability in practice. To solve the optimization problem involved in GSPL, a novel iterative algorithm is proposed with rigorously theoretical guarantees. Experimental results demonstrate the superb performance of GSPL on synthetic and real datasets.

scholarly journals A Fuzzy Co-Clustering Algorithm via Modularity Maximization

2018 ◽  
Vol 2018 ◽  
pp. 1-11
Author(s):  
Yongli Liu ◽  
Jingli Chen ◽  
Hao Chao
Keyword(s):  
Optimization Problem ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Experimental Studies ◽  
Optimization Procedure ◽  
Block Diagonal Matrix ◽  
Benchmark Datasets ◽  
Modularity Maximization ◽  
Object Feature ◽  
Block Diagonal

In this paper we propose a fuzzy co-clustering algorithm via modularity maximization, named MMFCC. In its objective function, we use the modularity measure as the criterion for co-clustering object-feature matrices. After converting into a constrained optimization problem, it is solved by an iterative alternative optimization procedure via modularity maximization. This algorithm offers some advantages such as directly producing a block diagonal matrix and interpretable description of resulting co-clusters, automatically determining the appropriate number of final co-clusters. The experimental studies on several benchmark datasets demonstrate that this algorithm can yield higher quality co-clusters than such competitors as some fuzzy co-clustering algorithms and crisp block-diagonal co-clustering algorithms, in terms of accuracy.

Gaussian Mixture Model Clustering with Incomplete Data

2021 ◽  
Vol 17 (1s) ◽  
pp. 1-14
Author(s):  
Yi Zhang ◽  
Miaomiao Li ◽  
Siwei Wang ◽  
Sisi Dai ◽  
Lei Luo ◽  
...  
Keyword(s):  
Gaussian Mixture Model ◽  
Mixture Model ◽  
Incomplete Data ◽  
Optimization Problem ◽  
Gaussian Mixture ◽  
Practical Applications ◽  
Learning Procedure ◽  
Benchmark Datasets ◽  
Effectiveness And Efficiency ◽  
Mixture Model Clustering

Gaussian mixture model (GMM) clustering has been extensively studied due to its effectiveness and efficiency. Though demonstrating promising performance in various applications, it cannot effectively address the absent features among data, which is not uncommon in practical applications. In this article, different from existing approaches that first impute the absence and then perform GMM clustering tasks on the imputed data, we propose to integrate the imputation and GMM clustering into a unified learning procedure. Specifically, the missing data is filled by the result of GMM clustering, and the imputed data is then taken for GMM clustering. These two steps alternatively negotiate with each other to achieve optimum. By this way, the imputed data can best serve for GMM clustering. A two-step alternative algorithm with proved convergence is carefully designed to solve the resultant optimization problem. Extensive experiments have been conducted on eight UCI benchmark datasets, and the results have validated the effectiveness of the proposed algorithm.

scholarly journals Localized Incomplete Multiple Kernel k-means

2018 ◽  
Author(s):  
Xinzhong Zhu ◽  
Xinwang Liu ◽  
Miaomiao Li ◽  
En Zhu ◽  
Li Liu ◽  
...  
Keyword(s):  
Iterative Algorithm ◽  
Local Structure ◽  
Optimization Problem ◽  
Clustering Algorithm ◽  
Recent Literature ◽  
State Of The Art ◽  
Nearest Neighbors ◽  
K Nearest Neighbors ◽  
Multiple Kernel ◽  
Benchmark Datasets

The recently proposed multiple kernel k-means with incomplete kernels (MKKM-IK) optimally integrates a group of pre-specified incomplete kernel matrices to improve clustering performance. Though it demonstrates promising performance in various applications, we observe that it does not \emph{sufficiently  consider the local structure among data and indiscriminately forces all pairwise sample similarity to equally align with their ideal similarity values}. This could make the incomplete kernels less effectively imputed, and in turn adversely affect the clustering performance. In this paper, we propose a novel localized incomplete multiple kernel k-means (LI-MKKM) algorithm to address this issue. Different from existing MKKM-IK, LI-MKKM only requires the similarity of a sample to its k-nearest neighbors to align with their ideal similarity values. This helps the clustering algorithm to focus on closer sample pairs that shall stay together and avoids involving unreliable similarity evaluation for farther sample pairs. We carefully design a three-step iterative algorithm to solve the resultant optimization problem and theoretically prove its convergence. Comprehensive experiments on eight benchmark datasets demonstrate that our algorithm significantly outperforms the state-of-the-art comparable algorithms proposed in the recent literature, verifying the advantage of considering local structure.

scholarly journals Incomplete Multiview Clustering via Late Fusion

2018 ◽  
Vol 2018 ◽  
pp. 1-11 ◽  
Author(s):  
Yongkai Ye ◽  
Xinwang Liu ◽  
Qiang Liu ◽  
Xifeng Guo ◽  
Jianping Yin
Keyword(s):  
Real World ◽  
Fusion Method ◽  
Multiple Views ◽  
Sensor Failure ◽  
Late Fusion ◽  
Imputation Methods ◽  
Early Fusion ◽  
Real World Applications ◽  
Benchmark Datasets ◽  
Multiview Clustering

In real-world applications of multiview clustering, some views may be incomplete due to noise, sensor failure, etc. Most existing studies in the field of incomplete multiview clustering have focused on early fusion strategies, for example, learning subspace from multiple views. However, these studies overlook the fact that clustering results with the visible instances in each view could be reliable under the random missing assumption; accordingly, it seems that learning a final clustering decision via late fusion of the clustering results from incomplete views would be more natural. To this end, we propose a late fusion method for incomplete multiview clustering. More specifically, the proposed method performs kernel k-means clustering on the visible instances in each view and then performs a late fusion of the clustering results from different views. In the late fusion step of the proposed method, we encode each view’s clustering result as a zero-one matrix, of which each row serves as a compressed representation of the corresponding instance. We then design an alternate updating algorithm to learn a unified clustering decision that can best group the visible compressed representations in each view according to the k-means clustering objective. We compare the proposed method with several commonly used imputation methods and a representative early fusion method on six benchmark datasets. The superior clustering performance observed validates the effectiveness of the proposed method.

scholarly journals Multi-View Multiple Clusterings Using Deep Matrix Factorization

2020 ◽  
Vol 34 (04) ◽  
pp. 6348-6355 ◽  
Author(s):  
Shaowei Wei ◽  
Jun Wang ◽  
Guoxian Yu ◽  
Carlotta Domeniconi ◽  
Xiangliang Zhang
Keyword(s):  
Matrix Factorization ◽  
State Of The Art ◽  
Optimization Procedure ◽  
Experimental Results ◽  
Layer By Layer ◽  
Complementary Information ◽  
Iterative Optimization ◽  
Multiple Clusterings ◽  
Benchmark Datasets ◽  
Data Matrices

Multi-view clustering aims at integrating complementary information from multiple heterogeneous views to improve clustering results. Existing multi-view clustering solutions can only output a single clustering of the data. Due to their multiplicity, multi-view data, can have different groupings that are reasonable and interesting from different perspectives. However, how to find multiple, meaningful, and diverse clustering results from multi-view data is still a rarely studied and challenging topic in multi-view clustering and multiple clusterings. In this paper, we introduce a deep matrix factorization based solution (DMClusts) to discover multiple clusterings. DMClusts gradually factorizes multi-view data matrices into representational subspaces layer-by-layer and generates one clustering in each layer. To enforce the diversity between generated clusterings, it minimizes a new redundancy quantification term derived from the proximity between samples in these subspaces. We further introduce an iterative optimization procedure to simultaneously seek multiple clusterings with quality and diversity. Experimental results on benchmark datasets confirm that DMClusts outperforms state-of-the-art multiple clustering solutions.

scholarly journals Least-thickness symmetric circular masonry arch of maximum horizontal thrust

2021 ◽  
Author(s):  
Giuseppe Cocchetti ◽  
Egidio Rizzi
Keyword(s):  
Optimization Problem ◽  
Research Work ◽  
Complementary Information ◽  
Masonry Arch ◽  
Collapse Mode ◽  
Hinge Position ◽  
Explicit Representations ◽  
Horizontal Thrust ◽  
Structural Optimization Problem ◽  
Typical Solution

AbstractThis analytical note shall provide a contribution to the understanding of general principles in the Mechanics of (symmetric circular) masonry arches. Within a mainstream of previous research work by the authors (and competent framing in the dedicated literature), devoted to investigate the classical structural optimization problem leading to the least-thickness condition under self-weight (“Couplet-Heyman problem”), and the relevant characteristics of the purely rotational five-hinge collapse mode, new and complementary information is here analytically derived. Peculiar extremal conditions are explicitly inspected, as those leading to the maximum intrinsic non-dimensional horizontal thrust and to the foremost wide angular inner-hinge position from the crown, both occurring for specific instances of over-complete (horseshoe) arches. The whole is obtained, and confronted, for three typical solution cases, i.e., Heyman, “CCR” and Milankovitch instances, all together, by full closed-form explicit representations, and elucidated by relevant illustrations.

Dynamic multi-objective matrix control for a class of switched systems

2021 ◽  
pp. 014233122199338
Author(s):  
Ali Thamallah ◽  
Anis Sakly ◽  
Faouzi M’Sahli
Keyword(s):  
Switched Systems ◽  
Optimization Problem ◽  
Control Method ◽  
Optimization Procedure ◽  
Control Law ◽  
Dynamic Matrix Control ◽  
General Reference ◽  
Multi Objective ◽  
Dynamic Matrix ◽  
Matrix Control

This article focuses on the tracking and stabilizing issues of a class of discrete switched systems. These systems are characterized by unknown switching sequences, a non-minimum phase, and time-varying or dead modes. In particular, for those governed by an indeterminate switching signal, it is very complicated to synthesize a control law able to systematically approach general reference-tracking difficulties. Taking into account the difficulty to express the dynamic of this class of systems, the present paper presents a new Dynamic matrix control method based on the multi-objective optimization and the truncated impulse response model. The formulation of the optimization problem aims to approach the general step-tracking issues under persistent and indeterminate mode changes and to overcome the stability problem along with retaining as many desirable features of the standard dynamic matrix control (DMC) method as possible. In addition, the formulated optimization problem integrates estimator variables able to manipulate the optimization procedure in favor of the active mode with an appropriate adjustment. It also provides a progressive and smooth multi-objective control law even in the presence of problems whether in subsystems or switching sequences. Finally, simulation examples and comparison tests are conducted to illustrate the potentiality and effectiveness of the developed method.

Optimization of the Robot Calibration Process

1992 ◽  
Author(s):  
G. Zak ◽  
R. G. Fenton ◽  
B. Benhabib
Keyword(s):  
Optimization Problem ◽  
Kinematic Model ◽  
Optimization Procedure ◽  
Calibration Procedure ◽  
Residual Error ◽  
Industrial Robots ◽  
Multiple Objective Optimization ◽  
Robot Calibration ◽  
Calibration Process ◽  
The Cost

Abstract Most industrial robots cannot be off-line programmed to carry out a task accurately, unless their kinematic model is suitably corrected through a calibration procedure. However, proper calibration is an expensive and time-consuming procedure due to the highly accurate measurement equipment required and due to the significant amount of data that must be collected. To improve the efficiency of robot calibration, an optimization procedure is proposed in this paper. The objective of minimizing the cost of the calibration is combined with the objective of minimizing the residual error after calibration in one multiple-objective optimization. Prediction of the residual error for a given calibration process presents the main difficulty for implementing the optimization. It is proposed that the residual error is expressed as a polynomial function. This function is obtained as a result of fitting a response surface to either experimental or simulated sample estimates of the residual error. The optimization problem is then solved by identifying a reduced set of possible solutions, thus greatly simplifying the decision maker’s choice of an effective calibration procedure. An application example of this method is also included.

scholarly journals Optimal design of a DC MHD pump by simulated annealing method

2014 ◽  
Vol 11 (2) ◽  
pp. 339-350
Author(s):  
Khadidja Bouali ◽  
Fatima Kadid ◽  
Rachid Abdessemed
Keyword(s):  
Simulated Annealing ◽  
Optimal Design ◽  
Objective Function ◽  
Design Problem ◽  
Design Methodology ◽  
Optimization Problem ◽  
Optimization Procedure ◽  
Simulated Annealing Method ◽  
Direct Interpretation ◽  
Annealing Method

In this paper a design methodology of a magnetohydrodynamic pump is proposed. The methodology is based on direct interpretation of the design problem as an optimization problem. The simulated annealing method is used for an optimal design of a DC MHD pump. The optimization procedure uses an objective function which can be the minimum of the mass. The constraints are both of geometrics and electromagnetic in type. The obtained results are reported.

Sign in / Sign up

Export Citation Format

Share Document