• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.1356-1358
• /
• 1996

Hopfield's suggestion of a neural network model for associative memory aroused the interest of many scientists and led to efforts of mathematical analyses. But the Hopfield Network has several disadvantages such as spurious states and capacity limitation. In that sense many scientists and engineers are trying to use a new optimization algorithm called genetic algorithm. But it is hard to use this algorithm in Hopfileld Network because of the fixed architecture. In this paper we introduce another method to determine the weight of Hopfield type network using Genetic Algorithm.

• 한국경영과학회지
• /
• 제17권1호
• /
• pp.3-16
• /
• 1992

We present a modified Tank and Hopfield's neural network model for solving Linear Programming problems. We have found the fact that the Tank and Hopfield's neural circuit for solving Linear Programming problems has some difficulties in guaranteeing convergence, and obtaining both the primal and dual optimum solutions from the output of the circuit. We have identified the exact conditions in which the circuit stops at an interior point of the feasible region, and therefore fails to converge. Also, proper scaling of the problem parameters is required, in order to obtain a feasible solution from the circuit. Even after one was successful in getting a primal optimum solution, the output of the circuit must be processed further to obtain a dual optimum solution. The modified model being proposed in the paper is designed to overcome such difficulties. We describe the modified model and summarize our computational experiment.

• 한국광학회:학술대회논문집
• /
• 한국광학회 1989년도 제4회 파동 및 레이저 학술발표회 4th Conference on Waves and lasers 논문집 - 한국광학회
• /
• pp.85-88
• /
• 1989

Introducing and optimizing it-significance to the Hopfield model, ten highly correlated binary images, i.e., numbers "0" to "9", are successfully stored and retrieved in a 6x8 node system. Unlike many other neural networks models, this model has stronger error correction capability for correlated images such as "6", "8", "3", and "9". the bit-significance optimization is regarded as an adaptive learning process based on least-mean-square error algorithm, and may be implemented with another neural nets optimizer. A design for electro-optic implementation including the adaptive optimization networks is also introduced.uding the adaptive optimization networks is also introduced.

• 한국컴퓨터정보학회논문지
• /
• 제4권4호
• /
• pp.94-100
• /
• 1999

본 논문은 흡필드 모델의 실수연산을 고속으로 수행할 수 있는 디지털 신경회로망의 구현에 관한 연구이다. 흡필드 모델［1］-［8］의 연산과정은 행렬-벡터의 연산으로 기술 할 수 있으며, 이 연산과정은 순환, 반복적으로 이루어지므로 어레이프로세서 구조로 설계하기에 적합하다. 또한, Look-up-Table(연산표)에 의하여 비선형 함수를 출력함으로써, 고속의 실수 연산을 수행할 수 있도록 설계하였다. 본 논문에서 제안한 방법은 현재 개발된 VLSI기술로 실현 가능하기 때문에 실제 신경회로망의 응용분야에 이용될 수 있을 것으로 기대된다.

• 한국측량학회지
• /
• 제31권6_2호
• /
• pp.559-565
• /
• 2013

Spatial resolution of land covers from remotely sensed images can be increased using super-resolution mapping techniques for soft-classified land cover proportions. A further development of super-resolution mapping technique is downscaling the original remotely sensed image using super-resolution mapping techniques with a forward model. In this paper, the model for increasing spatial resolution of remote sensing multispectral image is tested with real SPOT 5 imagery at 10m spatial resolution for an area in Bac Giang Province, Vietnam in order to evaluate the feasibility of application of this model to the real imagery. The soft-classified land cover proportions obtained using a fuzzy c-means classification are then used as input data for a Hopfield neural network (HNN) to predict the multispectral images at sub-pixel spatial resolution. The 10m SPOT multispectral image was improved to 5m, 3,3m and 2.5m and compared with SPOT Panchromatic image at 2.5m resolution for assessment.Visually, the resulted image is compared with a SPOT 5 panchromatic image acquired at the same time with the multispectral data. The predicted image is apparently sharper than the original coarse spatial resolution image.

• 전자공학회논문지B
• /
• 제30B권8호
• /
• pp.54-63
• /
• 1993

This paper proposes a method for improving the performances of the neural network for optimization by an optimal estimation of initial states. The optimal initial state that leads to the global minimum is estimated by using the stochastic approximation. And then the update rule of Hopfield model, which is the high speed deterministic algorithm using the steepest descent rule, is applied to speed up the optimization. The proposed method has been applied to the tavelling salesman problems and an optimal task partition problems to evaluate the performances. The simulation results show that the convergence speed of the proposed method is higher than conventinal Hopfield model. Abe's method and Boltzmann machine with random initial neuron output setting, and the convergence rate to the global minimum is guaranteed with probability of 1. The proposed method gives better result as the problem size increases where it is more difficult for the randomized initial setting to give a good convergence.

• 한국정밀공학회지
• /
• 제14권8호
• /
• pp.122-129
• /
• 1997

In this paper, we propose a Hopfield model for solving the part-matching in case that is the number of parts and positions are changed. The goal of this paper is to minimize part-connection in pairs and total path of part-connections. Therefore, this kind of problem is referred to as a combinatiorial optimization problem. First of all, we review the theoretical basis for Hopfield model and present two optimal algorithms of part-matching. The first algorithm is Traveling Salesman Problem(TSP) which improved the original and the second algorithm is Wdighted Matching Problem (WMP). Finally, we show demonstration through com- puter simulation and analyze the stability and feasibility of the generated solutions for the proposed con- nection methods. Therefore, we prove that the second algorithm is better than the first algorithm.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1995년도 춘계공동학술대회논문집; 전남대학교; 28-29 Apr. 1995
• /
• pp.600-600
• /
• 1995

• 전자공학회논문지C
• /
• 제36C권10호
• /
• pp.66-74
• /
• 1999

이 논문에서는 잘 알려진 N-P complete 문제인 TSP를 풀수 있는 통계적 홉필드 신경망의 일반화된 모델을 제안한다. 정규화를 통한 형태의 목적함수를 가진 반 덴 바우트의 방법은 필요한 외란 효과를 다 고려하지 않은 심각한 단점이 있다. 제안된 모델에서는 향상된 목적함수가 사용되었고 반 덴 바우트가 고려한 2가지와 박찬익이 더 고려한 1가지를 포함하는 5가지의 외란 효과와 외란 효과들의 비를 이용하는 방법을 제안한다. 임의로 만든 10개 도시의 시뮬레이션을 통해 제안한 모델이 100가지의 경우에서 90가지가 최적이나 거의 최적에 도달함을 보여준다. (오차 5% 이내로) 30개와 50개 도시의 큰 규모의 TSP에 대해 좋은 결과를 얻었다.

• 한국정보통신학회논문지
• /
• 제1권2호
• /
• pp.183-191
• /
• 1997

동시 다발적으로 공격해 오는 위협 표적을 방어하기는 매우 어려우며, 특히 방어용 무장수보다 표적의 수가 많을 경우에는 전체 표적 격추 기대 확률이 최대가 될 수 있도록 유지하는 방법으로서 본 논문에서는 홉필드 신경망 기법을 무장 할당 알고리즘으로 이용하는 방안을 제안하였다. 본 연구는 자동무장할당 알고리즘을 설계함에 있어서 할당변수를 생성하는데 필요한 신경망 학습 횟수를 단축하도록 설계하였으며 컴퓨터 시뮬레이션 결과 watcholder의 방법보다 수렴성이 뛰어남을 확인하였다.