• 한국지능시스템학회논문지
• /
• 제11권4호
• /
• pp.340-346
• /
• 2001

본 논문은 GBAM (general bidirectional associative memory) 모델을 위한 새로운 설계방법을 제시한다. GBAM 모델에 대한 이론적 고찰을 바탕으로, GBAM 기방 양방향 연상 메모리의 설계 문제가 GEVP (generalized eigenvalue problem)로 불리는 최적화 문제로 표현될 수 있음을 밝힌다. 설계 과정에서 등장하는 GEVP 문제들은 최근에 개발된 내부점 방법에 의하여 주어진 허용 오차 이내에서 효과적으로 풀릴 수 있으므로, 본 논문에서 확립된 설계 절차는 매우 실용적이다. 제안된 설계 절차에 대한 적용 가능성은 관련 연구에서 고려되었던 간단한 설계 예제를 통하여 예시된다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제1권1호
• /
• pp.35-43
• /
• 2001

In this paper, we propose a design method for GBSB (generalized brain-state-in-a-box) based associative memories. Based on the theoretical investigation about the properties of GBSB, we parameterize the solution space utilizing the limited number of parameters sufficient to represent the solution space and appropriate to be searched. Next we formulate the problem of finding a GBSB that can store the given pattern as stable states in the form of constrained optimization problems. Finally, we transform the constrained optimization problem into a SDP(semidefinite program), which can be solved by recently developed interior point methods. The applicability of the proposed method is illustrated via design examples.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.115.3-115
• /
• 2001

In this paper, discrete-time cellular neural networks are designed in order to function as associative memories by using Hebbian learning rule and non-cloning template. The proposed method has a very simple structure to design and to learn. Weights are updated by the connection between the neuron and its neighborhood. In the simulation, the proposed method is applied to the classification of a traffic sign pattern.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.1395-1397
• /
• 1996

Many of the existing neural associative memories are trained and recalled in separate modes and are not suitable for temporal pattern storage and classification in that user must specify the time and length of input patterns. In this paper, a new on-line temporal associative memory model is presented. This memory is structured in layers of neurons and each neuron has limited number of weights so that calculation complexity can be considerably reduced and processing of patterns can be achieved in real time.

• 생물정신의학
• /
• 제18권2호
• /
• pp.80-89
• /
• 2011

Pavlovian fear conditioning has been extensively studied for the understanding of neurobiological basis of memory and emotion. Pavlovian fear conditioning is an associative memory which forms when conditioned stimulus (CS) is paired with unconditioned stimulus (US) once or repeatedly. This behavioral model is also important for the understanding of anxiety disorders such as posttraumatic stress disorder. Here we describe the neural circuitry involved in fear conditioning and the molecular mechanisms underlying fear memory formation. During consolidation some memories fade out but other memories become stable and concrete. Emotion plays an important role in determining which memories will survive. Memory becomes unstable and editable again immediately after retrieval. It opens the possibility for us of modulating the established fear memory. It provides us with very efficient tools to improve the efficacy of cognitive-behavior therapy and other exposure-based therapy treating anxiety disorders.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.267-270
• /
• 1996

In this paper, we present a new design method to implement autoassociative memories based on BSB neural networks. With a concrete mathematical model proposed after analyzing some new qualitative properties of autoassociative memories, we reinterpret design of autoassociative memories as a constrained optimization problem and use an evolution program as an optimal search tool to solve this. The proposed method satisfies many of the criteria used to evaluate the effectivencess of a given associative memory and has improvements with respect to correctness and performance. Comparing simulation results with other methods, we demonstrate the effectiveness of the proposed method.

• 한국지능시스템학회논문지
• /
• 제10권2호
• /
• pp.161-167
• /
• 2000

본 논문에서는 양방향 연상 기능을 효과적으로 수행할 수 있는 BAM(bidirectional associative memory)의 설계방법론을 제안한다. 먼저 BAM의 성질에 관한 이론적 고찰을 바탕으로 하여 주어진 패턴 쌍을 안정하게 그리고 높은 오차수정율(error correction ratio)을 가지고 저장할 수 있는 BAM을 찾는 문제를 제약조건이 있는 최적화 문제로 공식화한다 다음과정에서 이 최적화 문제를 GEVP(generalized eigenvalue problem)로 변환함으로써 최근에 개발된 내부점 방법(interior point method)을 통하여 해가 구해질 수 있도록 한다. 제안된 설계 방법론의 적용가능성은 예제를 통해 확인된다.

• Transactions on Control, Automation and Systems Engineering
• /
• 제1권1호
• /
• pp.62-68
• /
• 1999

FCM(Fuzzy Cognitive Map) is proposed for representing causal reasoning. Its structure allows systematic causal reasoning through a forward inference. Authors have already proposed a diagnostic system based on FCM to utilized to identify the true origin of fault by on-line pattern diagnosis. In FCM based fault diagnosis, Temporal Associative Memories (TAM) recall of FCM is utilized to identify the true origin of fault by on-line pattern match where predicted pattern sequences obtained from TAM recall of fault FCM models are compared with actually observed ones. In engineering processes, the propagation delays are induced by the dynamics of processes and may vary with variables involved. However, disregarding such propagation delays in FCM-based fault diagnosis may lead to erroneous diagnostic results. To solve the problem, a concept of FTCM(Fuzzy Time Cognitive Map) is introduced into FCM-based fault diagnosis in this work. Expecially, translation method of FTCM makes it possible to diagnose the fault for some discrete time. Simulation studies through two-tank system is carried out to verify the effectiveness of the proposed diagnostic scheme.

• 한국정밀공학회지
• /
• 제12권6호
• /
• pp.64-71
• /
• 1995

The objective of this paper is to realize an analysis system that is capable of controlling the quality of an entire cutting process by including a 3 coordinate measuring machine in the process line. Fuzzy reasoning networks based on fuzzy associative memories has been intro- duced in the measuring process, the control limits for the control process have been obtained, and the efficiency and reliability of the system have been determined by examining the simu- lated reasoning control values.

• 한국지능시스템학회논문지
• /
• 제13권2호
• /
• pp.154-162
• /
• 2003

셀룰라 신경회로망의 연상 메모리를 이용하여 시각적인 입력 데이터의 연산을 통하여 영상 패턴의 분류와 인식을 수행한다. 셀룰라 신경회로망은 일반적인 신경회로망과 같이 비선형 데이터의 실시간 처리가 가능하고, 세포자동자와 같이 이 격자구조의 셀로 이루어져 인접한 셀과 직접 정보를 주고받는다. 응용 분야로는 최적화, 선형/비선형화, 연상 메모리, 패턴인식, 컴퓨터 비전 등에 적용할 수 있다. 영상의 이미지 픽셀을 셀룰라 신경회로망의 셀에 대응하여 전체 이미지 영상을 모든 셀룰라 신경회로망의 셀에서 동시에 병렬로 처리할 수 있어 2-D 이미지 처리에 적합하다. 본 논문은 셀룰라 신경회로망에 의한 연상 메모리 구조를 설계하고, 학습된 하중값 메모리에서 가장 적당한 하중값을 선택하여 학습된 영상과 정확히 일치하는 출력을 얻는 방법을 제시한다. 학습을 통한 연상 메모리 구현에는 각각의 뉴런에서 일정하지 않은 다른 템플릿을 사용한다. 각각의 템플릿은 뉴런들 간의 연결 하중값을 나타내고 학습에 따라 갱신된다. 학습방법으로는 템플릿 하중값 학습에 뉴런들 간의 연결 하중값을 조정하는 가장 단순한 규칙인 Hebb의 학습방법이 사용되었고 분류값 학습에 LMS 알고리즘이 사용되었다.