Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method
スポンサーリンク
概要
- 論文の詳細を見る
A genetic method to generate a neural network which has both structure and connection weights adequate for a given task is proposed. A neural network having arbitrary connections is regarded as a virtual living thing which has genes representing its connections among neural units. Effectiveness of the network is estimated from its time sequential input and output signals. Excellent individuals, namely appropriate neural networks, are generated through generation iterations. The basic principle of the method and its applications are described. As an example of evolution from randomly generated networks to feedforward networks, an XOR problem is dealt with, and an action control problem is used for making networks containing feedback and mutual connections. The proposed method is available for designing a neural network whose adequate structure is unknown.
- 社団法人電子情報通信学会の論文
- 1993-06-25
著者
-
Nagahashi Hiroshi
Department Of Information Processing Interdisciplinary Graduate School Of Science And Engineering To
-
Nagahashi Hiroshi
Imaging Science And Engineering Laboratory Tokyo Institute Of Technology
-
Agui Takeshi
Department Of Information Processing Interdisciplinary Graduate School Of Science And Engineering To
-
Agui Takeshi
Imaging Science And Engineering Laboratory Tokyo Institute Of Technology
-
Agui T
Toin Univ. Yokohama Yokohama‐shi Jpn
-
Nagahashi H
Tokyo Inst. Technol. Yokohama‐shi Jpn
-
Nagao Tomoharu
Imaging Science And Engineering Laboratory Tokyo Institute Of Technology
-
Nagao T
Graduate School Of Environment And Information Sciences Yokohama National University
関連論文
- Motion Segmentation in RGB Image Sequence Based on Stochastic Modeling
- Qコ-ダを用いた2値画像の可逆型順次符号化法〔英文〕
- Generating Breadth-First Expression for Gray Scale Quadtree
- A Method for C^2 Piecewise Quartic Polynomial Interpolation
- Quartic Interpolation on Triangles
- Piecewise Parametric Cubic Interpolation
- Representation of Surfaces on 5 and 6 Side Regions
- Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method
- A New Method for Smooth Interpolation without Twist Constraints
- DISTANCE MEASUREMENT OF A REAL-WORLD ENVIRONMENT USING AN ACTIVE CAMERA SYSTEM(International Workshop on Advanced Image Technology 2006)
- Interpolation of CT Slices for Laser Stereolithography
- Orientable Closed Surface Construction from Volume Data
- Pattern Recognition of Hand-Written Numerals by Round Field
- A Method for Watermarking to Bezier Polynomial Surface Models (Computer Graphics)
- Texture Classification Using Hierarchical Linear Discriminant Space(Image Recognition, Computer Vision)
- Parallel Computation of Parametric Piecewise Modeling Method
- COLOR RECOGNITION OF IMAGES USING A CATEGORICAL COLOR PERCEPTION MODEL(INTERNATIONAL Workshop on Advanced Image Technology 2008)
- A CATEGORICAL COLOR PERCEPTION MODEL USING STRUCTURED NEURAL NERWORK(International Workshop on Advanced Image Technology 2006)
- A Categorical Perception Model in consideration for Color Constancy (Medical Imaging and Computer Vision)(International Workshop On Advanced Image Technology (IWAIT2004))
- PT-ACTIT; Parameter Tunable - Automatic Construction of Tree-structural Image Transformation (Image Processing and Coding)(International Workshop On Advanced Image Technology (IWAIT2004))
- Texture-based Real-time Vision System for Detection of Automobile Cylinder Sleeves
- D-12-43 Photograph Based Pair-matching Recognition of Human Faces
- D-12-61 The pedestrian Tracking Based on an On-line Boosting Method
- Verification of physical properties of materials modeled with mass-spring systems (マルチメディア・仮想環境基礎)
- Verification of physical properties of materials modeled with mass-spring systems (画像工学)
- D-16-5 Multifractal Analysis for HCC Biopsy Images
- Classification of Prostate Histopathology Images Based on Multifractal Analysis
- D-12-22 A Simple Approach Tackling Illumination Problem of Face Detection
- Study on Illumination Insensitive Face Detection Based on Normalization Techniques