Norikazu Takahashi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Norikazu Takahashi has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
Tetsuo Nishi
Kyushu University, 744 Motooka, Nishi-ku, Fukuoka, 819-0395, Japan
10 shared publicationsJun Guo
7 shared publicationsT. Nishi
Dept. of Comput. Sci. & Commun. Eng., Kyushu Univ., Fukuoka, Japan
7 shared publicationsTsuyoshi Migita
Okayama University, 3-1-1 Tsuhima-naka, Kita-ku, Okayama, 700-8530, Japan
7 shared publicationsHajime Hara
Department of Electronics and Computer Engineering, Hiroshima Institute of Technology, Hiroshima 731-5193 Japan
3 shared publicationsTatsuya Fukami
Department of Informatics, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka, 819-0395, Japan
3 shared publications
Latest Publications
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- p. 1
A Novel Approximate Solution Method for Euclidean Steiner Tree Problem Based on Genetic Algorithm
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- p. 1
Open AccessA Novel Approximate Solution Method for Euclidean Steiner Tree Problem Based on Genetic Algorithm
- Open Access
- 12 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 82-86
Distributed Lasso Algorithm Based on Alternating Direction Method of Multipliers
- 12 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 82-86
Open AccessDistributed Lasso Algorithm Based on Alternating Direction Method of Multipliers
- Open Access
- 12 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 71-75
A Genetic Algorithm Approach to Euclidean Steiner Tree Problem
- 12 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 71-75
Open AccessA Genetic Algorithm Approach to Euclidean Steiner Tree Problem
- Open Access
- 1 July 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (7) , 1258-1261
Ensemble Learning in CNN Augmented with Fully Connected Subnetworks
Convolutional Neural Networks (CNNs) have shown remarkable performance in image recognition tasks. In this letter, we propose a new CNN model called the EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs). In this model, the set of feature maps generated by the last convolutional layer in the base CNN is divided along channels into disjoint subsets, and these subsets are assigned to the FCSNs. Each of the FCSNs is trained independent of others so that it can predict the class label of each feature map in the subset assigned to it. The output of the overall model is determined by majority vote of the base CNN and the FCSNs. Experimental results using the MNIST, Fashion-MNIST and CIFAR-10 datasets show that the proposed approach further improves the performance of CNNs. In particular, an EnsNet achieves a state-of-the-art error rate of 0.16% on MNIST.
- 1 July 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (7) , 1258-1261
Open AccessEnsemble Learning in CNN Augmented with Fully Connected Subnetworks
Convolutional Neural Networks (CNNs) have shown remarkable performance in image recognition tasks. In this letter, we propose a new CNN model called the EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs). In this model, the set of feature maps generated by the last convolutional layer in the base CNN is divided along channels into disjoint subsets, and these subsets are assigned to the FCSNs. Each of the FCSNs is trained independent of others so that it can predict the class label of each feature map in the subset assigned to it. The output of the overall model is determined by majority vote of the base CNN and the FCSNs. Experimental results using the MNIST, Fashion-MNIST and CIFAR-10 datasets show that the proposed approach further improves the performance of CNNs. In particular, an EnsNet achieves a state-of-the-art error rate of 0.16% on MNIST.
- Open Access
- 15 May 2023
- Published by Elsevier in Neural Networks
- Vol. 164, 588-605
Design of continuous-time recurrent neural networks with piecewise-linear activation function for generation of prescribed sequences of bipolar vectors
- 15 May 2023
- Published by Elsevier in Neural Networks
- Vol. 164, 588-605
Open AccessDesign of continuous-time recurrent neural networks with piecewise-linear activation function for generation of prescribed sequences of bipolar vectors
- Open Access
- 29 June 2022
- Published by Oxford University Press (OUP) in Journal of Complex Networks
- Vol. 10 (4)
An algorithm for updating betweenness centrality scores of all vertices in a graph upon deletion of a single edge
Betweenness centrality (BC) is a measure of the importance of a vertex in a graph, which is defined using the number of the shortest paths passing through the vertex. Brandes proposed an efficient algorithm for computing the BC scores of all vertices in a graph, which accumulates pair dependencies while traversing single-source shortest paths. Although this algorithm works well on static graphs, its direct application to dynamic graphs takes a huge amount of computation time because the BC scores must be computed from scratch every time the structure of graph changes. Therefore, various algorithms for updating the BC scores of all vertices have been developed so far. In this article, we propose a novel algorithm for updating the BC scores of all vertices in a graph upon deletion of a single edge. We also show the validity and efficiency of the proposed algorithm through theoretical analysis and experiments using various graphs obtained from synthetic and real networks.
- 29 June 2022
- Published by Oxford University Press (OUP) in Journal of Complex Networks
- Vol. 10 (4)
Open AccessAn algorithm for updating betweenness centrality scores of all vertices in a graph upon deletion of a single edge
Betweenness centrality (BC) is a measure of the importance of a vertex in a graph, which is defined using the number of the shortest paths passing through the vertex. Brandes proposed an efficient algorithm for computing the BC scores of all vertices in a graph, which accumulates pair dependencies while traversing single-source shortest paths. Although this algorithm works well on static graphs, its direct application to dynamic graphs takes a huge amount of computation time because the BC scores must be computed from scratch every time the structure of graph changes. Therefore, various algorithms for updating the BC scores of all vertices have been developed so far. In this article, we propose a novel algorithm for updating the BC scores of all vertices in a graph upon deletion of a single edge. We also show the validity and efficiency of the proposed algorithm through theoretical analysis and experiments using various graphs obtained from synthetic and real networks.
- Open Access
- 30 April 2022
- Published by Springer Nature in Journal of Global Optimization
- Vol. 84 (3) , 755-781
A novel update rule of HALS algorithm for nonnegative matrix factorization and Zangwill’s global convergence
Nonnegative Matrix Factorization (NMF) has attracted a great deal of attention as an effective technique for dimensionality reduction of large-scale nonnegative data. Given a nonnegative matrix, NMF aims to obtain two low-rank nonnegative factor matrices by solving a constrained optimization problem. The Hierarchical Alternating Least Squares (HALS) algorithm is a well-known and widely-used iterative method for solving such optimization problems. However, the original update rule used in the HALS algorithm is not well defined. In this paper, we propose a novel well-defined update rule of the HALS algorithm, and prove its global convergence in the sense of Zangwill. Unlike conventional globally-convergent update rules, the proposed one allows variables to take the value of zero and hence can obtain sparse factor matrices. We also present two stopping conditions that guarantee the finite termination of the HALS algorithm. The practical usefulness of the proposed update rule is shown through experiments using real-world datasets.
- 30 April 2022
- Published by Springer Nature in Journal of Global Optimization
- Vol. 84 (3) , 755-781
Open AccessA novel update rule of HALS algorithm for nonnegative matrix factorization and Zangwill’s global convergence
Nonnegative Matrix Factorization (NMF) has attracted a great deal of attention as an effective technique for dimensionality reduction of large-scale nonnegative data. Given a nonnegative matrix, NMF aims to obtain two low-rank nonnegative factor matrices by solving a constrained optimization problem. The Hierarchical Alternating Least Squares (HALS) algorithm is a well-known and widely-used iterative method for solving such optimization problems. However, the original update rule used in the HALS algorithm is not well defined. In this paper, we propose a novel well-defined update rule of the HALS algorithm, and prove its global convergence in the sense of Zangwill. Unlike conventional globally-convergent update rules, the proposed one allows variables to take the value of zero and hence can obtain sparse factor matrices. We also present two stopping conditions that guarantee the finite termination of the HALS algorithm. The practical usefulness of the proposed update rule is shown through experiments using real-world datasets.
- 17 November 2020
- Published by Springer Nature
- p. 342-350
Element-Wise Alternating Least Squares Algorithm for Nonnegative Matrix Factorization on One-Hot Encoded Data
- 17 November 2020
- Published by Springer Nature
- p. 342-350
Element-Wise Alternating Least Squares Algorithm for Nonnegative Matrix Factorization on One-Hot Encoded Data
- Open Access
Ensemble learning in CNN augmented with fully connected subnetworks
Convolutional Neural Networks (CNNs) have shown remarkable performance in general object recognition tasks. In this paper, we propose a new model called EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs). In this model, the set of feature-maps generated by the last convolutional layer in the base CNN is divided along channels into disjoint subsets, and these subsets are assigned to the FCSNs. Each of the FCSNs is trained independent of others so that it can predict the class label from the subset of the feature-maps assigned to it. The output of the overall model is determined by majority vote of the base CNN and the FCSNs. Experimental results using the MNIST, Fashion-MNIST and CIFAR-10 datasets show that the proposed approach further improves the performance of CNNs. In particular, an EnsNet achieves a state-of-the-art error rate of 0.16% on MNIST.
Open AccessEnsemble learning in CNN augmented with fully connected subnetworks
Convolutional Neural Networks (CNNs) have shown remarkable performance in general object recognition tasks. In this paper, we propose a new model called EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs). In this model, the set of feature-maps generated by the last convolutional layer in the base CNN is divided along channels into disjoint subsets, and these subsets are assigned to the FCSNs. Each of the FCSNs is trained independent of others so that it can predict the class label from the subset of the feature-maps assigned to it. The output of the overall model is determined by majority vote of the base CNN and the FCSNs. Experimental results using the MNIST, Fashion-MNIST and CIFAR-10 datasets show that the proposed approach further improves the performance of CNNs. In particular, an EnsNet achieves a state-of-the-art error rate of 0.16% on MNIST.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1332-1337
A Distributed HALS Algorithm for Euclidean Distance-Based Nonnegative Matrix Factorization
This paper proposes a distributed algorithm for multiple agents to perform the Nonnegative Matrix Factorization (NMF) based on the Euclidean distance. The matrix to be factorized is partitioned into multiple blocks, and each block is assigned to one of the agents forming a two-dimensional grid network. Each agent handles a small number of entries of the factor matrices corresponding to the assigned block, and updates their values by using information coming from the neighbors. It is shown that the proposed algorithm simulates the hierarchical alternating least squares method, which is well known as a fast algorithm for NMF based on the Euclidean distance, by making use of a finite-time distributed consensus algorithm.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1332-1337
A Distributed HALS Algorithm for Euclidean Distance-Based Nonnegative Matrix Factorization
This paper proposes a distributed algorithm for multiple agents to perform the Nonnegative Matrix Factorization (NMF) based on the Euclidean distance. The matrix to be factorized is partitioned into multiple blocks, and each block is assigned to one of the agents forming a two-dimensional grid network. Each agent handles a small number of entries of the factor matrices corresponding to the assigned block, and updates their values by using information coming from the neighbors. It is shown that the proposed algorithm simulates the hierarchical alternating least squares method, which is well known as a fast algorithm for NMF based on the Euclidean distance, by making use of a finite-time distributed consensus algorithm.