Kazuhisa Fujita
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
Research Subjects
Kazuhisa Fujita has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
Yoshiki Kashimori
Department of Engineering Science, University of Electro-Communications, Chofu, Japan
21 shared publicationsTakeshi Kambara
Department of Information Network Science, Graduate School of Information Systems, The University of Electro-Communications, Chofu, Tokyo, Japan
8 shared publications- Yoshiki Kashimori
University of Electro-Communications
5 shared publications MeiHong Zheng
Department of Applied Physics, School of Science and Engineering, Waseda University, Shinjuku, Tokyo 165-8555, Japan
2 shared publicationsShun Okuno
University of Electro-Communications
2 shared publicationsYuki Abe
University of Electro-Communications
2 shared publications
Latest Publications
- Open Access
- 28 November 2025
- Published by PeerJ in PeerJ Computer Science
- Vol. 11, e3403
AlphaViT: a flexible game-playing AI for multiple games and variable board sizes
This study presents three game-playing agents incorporating Vision Transformers (ViTs) into the AlphaZero framework: AlphaViT, AlphaViD (AlphaViT with a transformer decoder), and AlphaVDA (AlphaViD with learnable action embeddings). These agents can play multiple board games with varying sizes using a single shared-weight neural network, thus overcoming AlphaZero’s limitation of fixed board sizes. AlphaViT employs only a transformer encoder, whereas AlphaViD and AlphaVDA incorporate both a transformer encoder and a decoder. In AlphaViD, the decoder processes the output from the encoder, whereas AlphaVDA uses learnable embeddings as decoder input. The additional decoder in AlphaViD and AlphaVDA provides flexibility to adapt to various action spaces and board sizes. Experimental results show that the proposed agents, trained on either individual games or multiple games simultaneously, consistently outperform traditional algorithms, such as Minimax and Monte Carlo Tree Search. They achieve performance close to that of AlphaZero despite relying on a single deep neural network (DNN) with shared weights. In particular, AlphaViT performs well across all evaluated games. Furthermore, fine-tuning the DNN using weights pre-trained on small board games accelerates convergence and improves performance, particularly in Gomoku. Notably, simultaneous training on multiple games yields performance comparable to, or even surpassing, that of single-game training. These results indicate the potential of transformer-based architectures for developing flexible and robust game-playing AI agents that excel in multiple games and dynamic environments.
- 28 November 2025
- Published by PeerJ in PeerJ Computer Science
- Vol. 11, e3403
Open AccessAlphaViT: a flexible game-playing AI for multiple games and variable board sizes
This study presents three game-playing agents incorporating Vision Transformers (ViTs) into the AlphaZero framework: AlphaViT, AlphaViD (AlphaViT with a transformer decoder), and AlphaVDA (AlphaViD with learnable action embeddings). These agents can play multiple board games with varying sizes using a single shared-weight neural network, thus overcoming AlphaZero’s limitation of fixed board sizes. AlphaViT employs only a transformer encoder, whereas AlphaViD and AlphaVDA incorporate both a transformer encoder and a decoder. In AlphaViD, the decoder processes the output from the encoder, whereas AlphaVDA uses learnable embeddings as decoder input. The additional decoder in AlphaViD and AlphaVDA provides flexibility to adapt to various action spaces and board sizes. Experimental results show that the proposed agents, trained on either individual games or multiple games simultaneously, consistently outperform traditional algorithms, such as Minimax and Monte Carlo Tree Search. They achieve performance close to that of AlphaZero despite relying on a single deep neural network (DNN) with shared weights. In particular, AlphaViT performs well across all evaluated games. Furthermore, fine-tuning the DNN using weights pre-trained on small board games accelerates convergence and improves performance, particularly in Gomoku. Notably, simultaneous training on multiple games yields performance comparable to, or even surpassing, that of single-game training. These results indicate the potential of transformer-based architectures for developing flexible and robust game-playing AI agents that excel in multiple games and dynamic environments.
- Open Access
Flexible game-playing AI with AlphaViT: adapting to multiple games and board sizes
This paper presents novel game AI agents based on the AlphaZero framework, enhanced with Vision Transformers (ViT): AlphaViT, AlphaViD, and AlphaVDA. These agents are designed to play various board games of different sizes using a single model, overcoming AlphaZero's limitation of being restricted to a fixed board size. AlphaViT uses only a transformer encoder, while AlphaViD and AlphaVDA contain both an encoder and a decoder. AlphaViD's decoder receives input from the encoder output, while AlphaVDA uses a learnable matrix as decoder input. Using the AlphaZero framework, the three proposed methods demonstrate their versatility in different game environments, including Connect4, Gomoku, and Othello. Experimental results show that these agents, whether trained on a single game or on multiple games simultaneously, consistently outperform traditional algorithms such as Minimax and Monte Carlo tree search using a single DNN with shared weights, while approaching the performance of AlphaZero. In particular, AlphaViT and AlphaViD show strong performance across games, with AlphaViD benefiting from an additional decoder layer that enhances its ability to adapt to different action spaces and board sizes. These results may suggest the potential of transformer-based architectures to develop more flexible and robust game AI agents capable of excelling in multiple games and dynamic environments.
Open AccessFlexible game-playing AI with AlphaViT: adapting to multiple games and board sizes
This paper presents novel game AI agents based on the AlphaZero framework, enhanced with Vision Transformers (ViT): AlphaViT, AlphaViD, and AlphaVDA. These agents are designed to play various board games of different sizes using a single model, overcoming AlphaZero's limitation of being restricted to a fixed board size. AlphaViT uses only a transformer encoder, while AlphaViD and AlphaVDA contain both an encoder and a decoder. AlphaViD's decoder receives input from the encoder output, while AlphaVDA uses a learnable matrix as decoder input. Using the AlphaZero framework, the three proposed methods demonstrate their versatility in different game environments, including Connect4, Gomoku, and Othello. Experimental results show that these agents, whether trained on a single game or on multiple games simultaneously, consistently outperform traditional algorithms such as Minimax and Monte Carlo tree search using a single DNN with shared weights, while approaching the performance of AlphaZero. In particular, AlphaViT and AlphaViD show strong performance across games, with AlphaViD benefiting from an additional decoder layer that enhances its ability to adapt to different action spaces and board sizes. These results may suggest the potential of transformer-based architectures to develop more flexible and robust game AI agents capable of excelling in multiple games and dynamic environments.
- Open Access
- 8 January 2024
- Published by PeerJ in PeerJ Computer Science
- Vol. 10, e1789
An efficient and straightforward online vector quantization method for a data stream through remove-birth updating
The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This article proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.
- 8 January 2024
- Published by PeerJ in PeerJ Computer Science
- Vol. 10, e1789
Open AccessAn efficient and straightforward online vector quantization method for a data stream through remove-birth updating
The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This article proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.
- Open Access
An efficient and straightforward online quantization method for a data stream through remove-birth updating
The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This paper proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.
Open AccessAn efficient and straightforward online quantization method for a data stream through remove-birth updating
The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This paper proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.
- Open Access
Characteristics of networks generated by kernel growing neural gas
This research aims to develop kernel GNG, a kernelized version of the growing neural gas (GNG) algorithm, and to investigate the features of the networks generated by the kernel GNG. The GNG is an unsupervised artificial neural network that can transform a dataset into an undirected graph, thereby extracting the features of the dataset as a graph. The GNG is widely used in vector quantization, clustering, and 3D graphics. Kernel methods are often used to map a dataset to feature space, with support vector machines being the most prominent application. This paper introduces the kernel GNG approach and explores the characteristics of the networks generated by kernel GNG. Five kernels, including Gaussian, Laplacian, Cauchy, inverse multiquadric, and log kernels, are used in this study. The results of this study show that the average degree and the average clustering coefficient decrease as the kernel parameter increases for Gaussian, Laplacian, Cauchy, and IMQ kernels. If we avoid more edges and a higher clustering coefficient (or more triangles), the kernel GNG with a larger value of the parameter will be more appropriate.
Open AccessCharacteristics of networks generated by kernel growing neural gas
This research aims to develop kernel GNG, a kernelized version of the growing neural gas (GNG) algorithm, and to investigate the features of the networks generated by the kernel GNG. The GNG is an unsupervised artificial neural network that can transform a dataset into an undirected graph, thereby extracting the features of the dataset as a graph. The GNG is widely used in vector quantization, clustering, and 3D graphics. Kernel methods are often used to map a dataset to feature space, with support vector machines being the most prominent application. This paper introduces the kernel GNG approach and explores the characteristics of the networks generated by kernel GNG. Five kernels, including Gaussian, Laplacian, Cauchy, inverse multiquadric, and log kernels, are used in this study. The results of this study show that the average degree and the average clustering coefficient decrease as the kernel parameter increases for Gaussian, Laplacian, Cauchy, and IMQ kernels. If we avoid more edges and a higher clustering coefficient (or more triangles), the kernel GNG with a larger value of the parameter will be more appropriate.
- Open Access
- 4 October 2022
- Published by PeerJ in PeerJ Computer Science
- Vol. 8, e1123
AlphaDDA: strategies for adjusting the playing strength of a fully trained AlphaZero system to a suitable human training partner
Artificial intelligence (AI) has achieved superhuman performance in board games such as Go, chess, and Othello (Reversi). In other words, the AI system surpasses the level of a strong human expert player in such games. In this context, it is difficult for a human player to enjoy playing the games with the AI. To keep human players entertained and immersed in a game, the AI is required to dynamically balance its skill with that of the human player. To address this issue, we propose AlphaDDA, an AlphaZero-based AI with dynamic difficulty adjustment (DDA). AlphaDDA consists of a deep neural network (DNN) and a Monte Carlo tree search, as in AlphaZero. AlphaDDA learns and plays a game the same way as AlphaZero, but can change its skills. AlphaDDA estimates the value of the game state from only the board state using the DNN. AlphaDDA changes a parameter dominantly controlling its skills according to the estimated value. Consequently, AlphaDDA adjusts its skills according to a game state. AlphaDDA can adjust its skill using only the state of a game without any prior knowledge regarding an opponent. In this study, AlphaDDA plays Connect4, Othello, and 6x6 Othello with other AI agents. Other AI agents are AlphaZero, Monte Carlo tree search, the minimax algorithm, and a random player. This study shows that AlphaDDA can balance its skill with that of the other AI agents, except for a random player. AlphaDDA can weaken itself according to the estimated value. However, AlphaDDA beats the random player because AlphaDDA is stronger than a random player even if AlphaDDA weakens itself to the limit. The DDA ability of AlphaDDA is based on an accurate estimation of the value from the state of a game. We believe that the AlphaDDA approach for DDA can be used for any game AI system if the DNN can accurately estimate the value of the game state and we know a parameter controlling the skills of the AI system.
- 4 October 2022
- Published by PeerJ in PeerJ Computer Science
- Vol. 8, e1123
Open AccessAlphaDDA: strategies for adjusting the playing strength of a fully trained AlphaZero system to a suitable human training partner
Artificial intelligence (AI) has achieved superhuman performance in board games such as Go, chess, and Othello (Reversi). In other words, the AI system surpasses the level of a strong human expert player in such games. In this context, it is difficult for a human player to enjoy playing the games with the AI. To keep human players entertained and immersed in a game, the AI is required to dynamically balance its skill with that of the human player. To address this issue, we propose AlphaDDA, an AlphaZero-based AI with dynamic difficulty adjustment (DDA). AlphaDDA consists of a deep neural network (DNN) and a Monte Carlo tree search, as in AlphaZero. AlphaDDA learns and plays a game the same way as AlphaZero, but can change its skills. AlphaDDA estimates the value of the game state from only the board state using the DNN. AlphaDDA changes a parameter dominantly controlling its skills according to the estimated value. Consequently, AlphaDDA adjusts its skills according to a game state. AlphaDDA can adjust its skill using only the state of a game without any prior knowledge regarding an opponent. In this study, AlphaDDA plays Connect4, Othello, and 6x6 Othello with other AI agents. Other AI agents are AlphaZero, Monte Carlo tree search, the minimax algorithm, and a random player. This study shows that AlphaDDA can balance its skill with that of the other AI agents, except for a random player. AlphaDDA can weaken itself according to the estimated value. However, AlphaDDA beats the random player because AlphaDDA is stronger than a random player even if AlphaDDA weakens itself to the limit. The DDA ability of AlphaDDA is based on an accurate estimation of the value from the state of a game. We believe that the AlphaDDA approach for DDA can be used for any game AI system if the DNN can accurately estimate the value of the game state and we know a parameter controlling the skills of the AI system.
- Open Access
AlphaDDA: Strategies for Adjusting the Playing Strength of a Fully Trained AlphaZero System to a Suitable Human Training Partner
Artificial intelligence (AI) has achieved superhuman performance in board games such as Go, chess, and Othello (Reversi). In other words, the AI system surpasses the level of a strong human expert player in such games. In this context, it is difficult for a human player to enjoy playing the games with the AI. To keep human players entertained and immersed in a game, the AI is required to dynamically balance its skill with that of the human player. To address this issue, we propose AlphaDDA, an AlphaZero-based AI with dynamic difficulty adjustment (DDA). AlphaDDA consists of a deep neural network (DNN) and a Monte Carlo tree search, as in AlphaZero. AlphaDDA learns and plays a game the same way as AlphaZero, but can change its skills. AlphaDDA estimates the value of the game state from only the board state using the DNN. AlphaDDA changes a parameter dominantly controlling its skills according to the estimated value. Consequently, AlphaDDA adjusts its skills according to a game state. AlphaDDA can adjust its skill using only the state of a game without any prior knowledge regarding an opponent. In this study, AlphaDDA plays Connect4, Othello, and 6x6 Othello with other AI agents. Other AI agents are AlphaZero, Monte Carlo tree search, the minimax algorithm, and a random player. This study shows that AlphaDDA can balance its skill with that of the other AI agents, except for a random player. The DDA ability of AlphaDDA is based on an accurate estimation of the value from the state of a game. We believe that the AlphaDDA approach for DDA can be used for any game AI system if the DNN can accurately estimate the value of the game state and we know a parameter controlling the skills of the AI system.
Open AccessAlphaDDA: Strategies for Adjusting the Playing Strength of a Fully Trained AlphaZero System to a Suitable Human Training Partner
Artificial intelligence (AI) has achieved superhuman performance in board games such as Go, chess, and Othello (Reversi). In other words, the AI system surpasses the level of a strong human expert player in such games. In this context, it is difficult for a human player to enjoy playing the games with the AI. To keep human players entertained and immersed in a game, the AI is required to dynamically balance its skill with that of the human player. To address this issue, we propose AlphaDDA, an AlphaZero-based AI with dynamic difficulty adjustment (DDA). AlphaDDA consists of a deep neural network (DNN) and a Monte Carlo tree search, as in AlphaZero. AlphaDDA learns and plays a game the same way as AlphaZero, but can change its skills. AlphaDDA estimates the value of the game state from only the board state using the DNN. AlphaDDA changes a parameter dominantly controlling its skills according to the estimated value. Consequently, AlphaDDA adjusts its skills according to a game state. AlphaDDA can adjust its skill using only the state of a game without any prior knowledge regarding an opponent. In this study, AlphaDDA plays Connect4, Othello, and 6x6 Othello with other AI agents. Other AI agents are AlphaZero, Monte Carlo tree search, the minimax algorithm, and a random player. This study shows that AlphaDDA can balance its skill with that of the other AI agents, except for a random player. The DDA ability of AlphaDDA is based on an accurate estimation of the value from the state of a game. We believe that the AlphaDDA approach for DDA can be used for any game AI system if the DNN can accurately estimate the value of the game state and we know a parameter controlling the skills of the AI system.
- Open Access
Extract an essential skeleton of a character as a graph from a character image
This paper aims to make a graph representing an essential skeleton of a character from an image that includes a machine printed or a handwritten character using growing neural gas (GNG) method and relative network graph (RNG) algorithm. The visual system in our brain can recognize printed characters and handwritten characters easily, robustly, and precisely. How does our brain robustly recognize characters? The visual processing in our brain uses the essential features of an object, such as crosses and corners. These features will be helpful for character recognition by a computer. However, extraction of the features is difficult. If the skeleton of a character is represented as a graph, we can more easily extract the features. To extract the skeleton of a character as a graph from an image, this paper proposes the new approach using GNG and RNG algorithm. I achieved to extract skeleton graphs from images including distorted, noisy, and handwritten characters.
Open AccessExtract an essential skeleton of a character as a graph from a character image
This paper aims to make a graph representing an essential skeleton of a character from an image that includes a machine printed or a handwritten character using growing neural gas (GNG) method and relative network graph (RNG) algorithm. The visual system in our brain can recognize printed characters and handwritten characters easily, robustly, and precisely. How does our brain robustly recognize characters? The visual processing in our brain uses the essential features of an object, such as crosses and corners. These features will be helpful for character recognition by a computer. However, extraction of the features is difficult. If the skeleton of a character is represented as a graph, we can more easily extract the features. To extract the skeleton of a character as a graph from an image, this paper proposes the new approach using GNG and RNG algorithm. I achieved to extract skeleton graphs from images including distorted, noisy, and handwritten characters.
- Open Access
- 20 August 2021
- Published by PeerJ in PeerJ Computer Science
- Vol. 7, e679
Approximate spectral clustering using both reference vectors and topology of the network generated by growing neural gas
Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.
- 20 August 2021
- Published by PeerJ in PeerJ Computer Science
- Vol. 7, e679
Open AccessApproximate spectral clustering using both reference vectors and topology of the network generated by growing neural gas
Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.
- Open Access
Approximate spectral clustering using both reference vectors and topology of the network generated by growing neural gas
Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.
Open AccessApproximate spectral clustering using both reference vectors and topology of the network generated by growing neural gas
Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.