The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
지형 구성의 메커니즘을 규명하기 위해 일반화된 세포층에서 일반화된 세포층으로의 간단한 지형 매핑 형성 모델을 제안합니다. 여기서 일반화된 셀 계층은 임의의 셀 이웃 관계를 고려한다는 의미입니다. 이전 연구에서 우리는 1차원 세포층 사이의 지형 매핑 형성 모델을 조사했습니다. 이 논문에서는 세포층 구조를 모든 차원으로 확장합니다. 우리 모델에서 각 셀은 이진 상태 값을 취하고 Hebb의 규칙과 Anti-Hebb의 규칙을 확장한 학습 원리 클래스를 고려합니다. 우리는 시냅스 전과 후의 세포 상태가 동일한 값을 가질 경우 시냅스 가중치 값이 증가하는 상관형 학습 규칙에 특히 주의합니다. 먼저 반임베딩인 경우에만 상관 학습과 관련하여 매핑이 안정적이라는 것을 보여줍니다. 둘째, 우리는 밴드 유형(band type)이라고 불리는 특별한 클래스의 가중치 행렬을 소개하고 밴드 유형 가중치 행렬 세트가 강하게 닫혀 있으며 이러한 가중치 행렬은 지형 매핑을 생성할 수 없음을 보여줍니다. 셋째, 매핑이 밴드형이 아닌 가중치 행렬로 정의되면 상관 학습 규칙에 따라 지형 매핑으로 수렴하는 것을 컴퓨터 시뮬레이션을 통해 보여줍니다.
지형 매핑, 안정, Hebbian 학습, 비지도 학습, 상관 학습
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Shouji SAKAMOTO, Youichi KOBUCHI, "Stability of Topographic Mappings between Generalized Cell Layers" in IEICE TRANSACTIONS on Information,
vol. E85-D, no. 7, pp. 1145-1152, July 2002, doi: .
Abstract: To elucidate the mechanism of topographic organization, we propose a simple topographic mapping formation model from generalized cell layer to generalized cell layer. Here generalized cell layer means that we consider arbitrary cell neighborhood relations. In our previous work we investigated a topographic mapping formation model between one dimensional cell layers. In this paper we extend the cell layer structure to any dimension. In our model, each cell takes a binary state value and we consider a class of learning principles which are extensions of Hebb's rule and Anti-Hebb's rule. We pay special attention to correlation type learning rules where a synaptic weight value is increased if pre and post synaptic cell states have the same value. We first show that a mapping is stable with respect to the correlational learning if and only if it is semi-embedding. Second, we introduce a special class of weight matrices called band type and show that the set of band type weight matrices is strongly closed and such a weight matrix can not yield a topographic mapping. Third, we show by computer simulations that a mapping, if it is defined by a non band type weight matrix, converges to a topographic mapping under the correlational learning rules.
URL: https://global.ieice.org/en_transactions/information/10.1587/e85-d_7_1145/_p
부
@ARTICLE{e85-d_7_1145,
author={Shouji SAKAMOTO, Youichi KOBUCHI, },
journal={IEICE TRANSACTIONS on Information},
title={Stability of Topographic Mappings between Generalized Cell Layers},
year={2002},
volume={E85-D},
number={7},
pages={1145-1152},
abstract={To elucidate the mechanism of topographic organization, we propose a simple topographic mapping formation model from generalized cell layer to generalized cell layer. Here generalized cell layer means that we consider arbitrary cell neighborhood relations. In our previous work we investigated a topographic mapping formation model between one dimensional cell layers. In this paper we extend the cell layer structure to any dimension. In our model, each cell takes a binary state value and we consider a class of learning principles which are extensions of Hebb's rule and Anti-Hebb's rule. We pay special attention to correlation type learning rules where a synaptic weight value is increased if pre and post synaptic cell states have the same value. We first show that a mapping is stable with respect to the correlational learning if and only if it is semi-embedding. Second, we introduce a special class of weight matrices called band type and show that the set of band type weight matrices is strongly closed and such a weight matrix can not yield a topographic mapping. Third, we show by computer simulations that a mapping, if it is defined by a non band type weight matrix, converges to a topographic mapping under the correlational learning rules.},
keywords={},
doi={},
ISSN={},
month={July},}
부
TY - JOUR
TI - Stability of Topographic Mappings between Generalized Cell Layers
T2 - IEICE TRANSACTIONS on Information
SP - 1145
EP - 1152
AU - Shouji SAKAMOTO
AU - Youichi KOBUCHI
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E85-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2002
AB - To elucidate the mechanism of topographic organization, we propose a simple topographic mapping formation model from generalized cell layer to generalized cell layer. Here generalized cell layer means that we consider arbitrary cell neighborhood relations. In our previous work we investigated a topographic mapping formation model between one dimensional cell layers. In this paper we extend the cell layer structure to any dimension. In our model, each cell takes a binary state value and we consider a class of learning principles which are extensions of Hebb's rule and Anti-Hebb's rule. We pay special attention to correlation type learning rules where a synaptic weight value is increased if pre and post synaptic cell states have the same value. We first show that a mapping is stable with respect to the correlational learning if and only if it is semi-embedding. Second, we introduce a special class of weight matrices called band type and show that the set of band type weight matrices is strongly closed and such a weight matrix can not yield a topographic mapping. Third, we show by computer simulations that a mapping, if it is defined by a non band type weight matrix, converges to a topographic mapping under the correlational learning rules.
ER -