http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Self-Deleting Neural Network for Vector Quantization
Maeda, Michiharu,Miyajima, Hiromi,Murashima, Sadayuki 대한전자공학회 1996 APCCAS:Asia Pacific Conference on Circuits And Sys Vol.1 No.1
Vector quantization is required the algorithm that minimizes the distortion error, and used for both storage and transmission of speech and image data. For a neural vector quantization [1], the self-creating neural network [2] and self-deleting neural network [3] and known for showing fine characters. In this paper, we improve the self-deleting neural network, and propose a generalization algorithm combining the creating and deleting neural networks. We discuss algorithms with neighborhood relations [2]∼[5] compared with the proposed one. Experimental results show the effectiveness of the proposed algorithm.
Some Properties of Quantum Data Search Algorithms
Keisuke Arima,Hiromi Miyajima,Noritaka Shigei,Michiharu Maeda 대한전자공학회 2008 ITC-CSCC :International Technical Conference on Ci Vol.2008 No.7
This paper deals with some properties of quantum data search algorithms. First, Grover’s and Ventura’s algorithms for quantum data search are introduced and compared with each other. As a result, it is shown that both algorithms are not always universal with the number of stored data. Further, some properties on the data search algorithms are shown.
A Learning Algorithm of Self-Organizing Maps for Image Restoration
Michiharu Maeda,Hiromi Miyajima 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
This paper presents a learning algorithm of selforganizing maps for image restoration. Our algorithm creates a map containing one unit for each pixel. Utilizing pixel values as input, image inference is conducted by selforganizing maps. Experimental results are presented in order to show that our approach is effective in quality.
Kazuya Kishida,Shoichi Nagata,Hiromi Miyajima 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
In this research, we propose the construction method of multi-agent systems for capturing fleeing targets using genetic algorithms. In order to construct a simplified multi-agent system, we introduce gray zones to an agent’s neighboring area by reducing input. We demonstrate the effectiveness of a proposed method using some numerical simulations.
Fuzzy Inference Models Appropriate for Digital Circuit
Shinya Nagamine,Noritaka Shigei,Hiromi Miyajima 대한전자공학회 2008 ITC-CSCC :International Technical Conference on Ci Vol.2008 No.7
In this paper, we propose fuzzy learning models suitable for digital circuit implementation. In the first place, we propose a division-free model which is advantageous in the circuit size and the processing speed. Next, we also consider to improve the accuracy of the division-free model by using ensemble learning. The effectiveness of the proposed methods is demonstrated by numerical simulations. Further, we describe a digital circuit design of the proposed model, and show the effectiveness of the design by the implementation on FPGA.