• 지능정보연구
• /
• 제10권2호
• /
• pp.53-63
• /
• 2004

순환 케스케이드 코릴레이션(Recurrent Cascade Correlation(RCC))은 감독에 의하여 학습하는 알고리즘이고 네트워크의 크기와 형태는 자동으로 이루어진다. RCC는 새로운 은닉뉴런들이 한 충에 하나씩 순서대로 네트워크에 삽입되기 때문에 다층구조를 형성하고 2계(Second Order) RCC는 새로운 은닉뉴런들이 한 층에만 순서대로 생성되어 나열되므로 2층 구조를 형성한다. 따라서 이러한 은닉뉴런들끼리는 서로 연결하지 않는다. 이 논문에서는 RCC와 2계 RCC의 조합을 통한 RCC 네트워크의 일반화를 소개한다. 새로운 RCC 알고리즘은 은닉뉴런이 네트워크에 첨가될 때마다 네트워크가 수직성장 또는 수평성장을 해야 하는지를 스스로 결정한다. 또한 뉴런의 활성화를 위한 새로운 활성화함수를 소개하고 기존의 sigmoid, tanh 함수와 함께 사용하여 모스 벤치마크 문제에 관하여 실험하였다. 이러한 활성화 함수들을 사용한 RCC 네트워크의 일반화 실험에서 은닉뉴런의 숫자가 감소하였음을 알 수 있다.

• 조명전기설비학회논문지
• /
• 제26권9호
• /
• pp.73-79
• /
• 2012

In this paper, we study on how to determine the number of hidden layer neurons in neural network for predicting defect size of steam generator tube. It was reported in the literature that the number of hidden layer neurons can be efficiently determined with the help of cross-validation. Although the cross-validation provides decent estimation performance in most cases, the performance depends on the selection of validation set and rather poor performance may be led to in some cases. In order to avoid such a problem, we propose to use multifold cross-validation. Through the simulation study, it is shown that the estimation performance of defect width (defect depth, respectively) attains 94% (99.4%, respectively) of the best performance achievable among the considered neuron numbers.

• 조명전기설비학회논문지
• /
• 제24권1호
• /
• pp.63-70
• /
• 2010

본 논문에서는 원자력발전소 증기세관 크기 예측을 위한 신경회로망 구조에 대해서 연구한다. 와류탐상 시험(ECT) 신호로부터 특징을 추출한 후, 결함크기 예측을 위해서 다층퍼셉트론 신경회로망을 이용한다. 결함크기 예측성능을 최대화하기 위해서는 신경회로망의 구조, 특히 은닉층 내의 뉴런의 개수를 신중히 결정하여야 한다. 본 논문에서는, 결함크기 예측을 위한 은닉층 내의 뉴런의 개수를 교차검증을 이용하여 매우 효과적으로 결정할 수 있음을 보인다.

• Journal of Korean Biological Nursing Science
• /
• 제24권2호
• /
• pp.77-85
• /
• 2022

Purpose: Aging process comes with cognitive impairment due to decreased neuronal cell number, activity, and neuronal circuit. Alteration of inhibitory neurons contributes to cognitive impairment in normal aging and is responsible for disrupting the excitation/inhibition balance by reducing the synthesis of gamma-aminobutyric acid (GABA). Morus nigra (Mulberry) is a natural physiologically active substance that has been proven to have anti-oxidant, anti-diabetic, and anti-inflammatory effects through many studies. This study aimed to evaluate the effects of the mulberry extract (ME) on cognitive function through anti-oxidant enzyme and GABAergic neuronal activity in aged rat brain. Methods: Sprague Dawley rats were randomly assigned as the young group (8 weeks, n= 8), aging group (67 weeks, n= 8), and aging+ mulberry extract group (67 weeks, n= 8). The aging+ mulberry extract group was orally administered 500 mg/kg/d mulberry extract for 6 weeks. Results: The aging+ mulberry extract group improved spatial and short-term memory. The antioxidant potential of ME increased the expression of superoxide dismutase-1 (SOD-1) and decreased inducible nitric oxide synthase (iNOS). Also, the aging+ mulberry extract group significantly increased the expression of GABAergic interneuron in hippocampus cornu ammonis1 (CA1) compared to the aging group. Conclusion: The number of GABAergic inhibitory interneurons was deceased and memory functions in the aging process, but those symptoms were improved and restored by mulberry extract administration.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.451-455
• /
• 2009

This paper provides a comparison of global path planning method in single string by using pulled and pushed SOFM (Self-Organizing Feature Map) which is a method among a number of neural network. The self-organizing feature map uses a randomized small valued initial-weight-vectors, selects the neuron whose weight vector best matches input as the winning neuron, and trains the weight vectors such that neurons within the activity bubble are move toward the input vector. On the other hand, the modified SOFM method in this research uses a predetermined initial weight vectors of the one dimensional string, gives the systematic input vector whose position best matches obstacles, and trains the weight vectors such that neurons within the activity bubble are move toward or reverse the input vector, by rising a pulled- or a pushed-SOFM. According to simulation results one can conclude that the modified neural networks in single string are useful tool for the global path planning problem of a mobile robot. In comparison of the number of iteration for converging to the solution the pushed-SOFM is more useful than the pulled-SOFM in global path planning for mobile robot.

• Molecules and Cells
• /
• 제27권4호
• /
• pp.397-401
• /
• 2009

Olig2 transcription factor is widely expressed throughout the central nervous system; therefore, it is considered to have multiple functions in the developing, mature and injured brain. In this mini-review, we focus on Olig2 in the forebrain (telencephalon and diencephalon) and discuss the functional significance of Olig2 and the differentiation properties of Olig2-expressing progenitors in the development and injured states. Short- and long-term lineage analysis in the developing forebrain elucidated that not all late Olig2+ cells are direct cohorts of early cells and that Olig2 lineage cells differentiate into neurons or glial cells in a region- and stage-dependent manner. Olig2-deficient mice revealed large elimination of oligodendrocyte precursor cells and a decreased number of astrocyte progenitors in the dorsal cortex, whereas no reduction in the number of GABAergic neurons. In addition to Olig2 function in the developing cortex, Olig2 is also reported to be important for glial scar formation after injury. Thus, Olig2 can be essential for glial differentiation during development and after injury.

• 융합신호처리학회논문지
• /
• 제11권3호
• /
• pp.250-254
• /
• 2010

As the number of input and fuzzy set of a fuzzy system increase, the size of the rule base increases exponentially and becomes unmanageable (curse of dimensionality). In this paper, tree architectures of fuzzy controller (TAFC) is proposed to overcome the curse of dimensionality problem occurring in the design of fuzzy controller. TAFC is constructed with the aid of AND and OR fuzzy neurons. TAFC can guarantee reduced size of rule base with reasonable performance. For the development of TAFC, genetic algorithm constructs the binary tree structure by optimally selecting the nodes and leaves, and then random signal-based learning further refines the binary connections (two-step optimization). An inverted pendulum system is considered to verify the effectiveness of the proposed method by simulation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.281-286
• /
• 2005

The main objective of the research work presented in this article is to present a systematic approach for designing a multilayer feed-forward artificial neural network based self-tuning power system stabilizer (ST-ANNPSS). In order to suggest an approach for selecting the number of neurons in the hidden layer, the dynamic performance of the system with ST-ANNPSS is studied and hence compared with that of conventional PSS. Finally the effect of variation of loading condition and equivalent reactance, Xe is investigated on dynamic performance of the system with ST-ANNPSS. Investigations reveal that ANN with one hidden layer comprising nine neurons is adequate and sufficient for ST-ANNPSS. Studies show that the dynamic performance of STANNPSS is quite superior to that of conventional PSS for the loading condition different from the nominal. Also it is revealed that the performance of ST-ANNPSS is quite robust to a wide variation in loading condition.

• Bulletin of the Korean Chemical Society
• /
• 제26권3호
• /
• pp.399-408
• /
• 2005

We have utilized neural network (NN) studies to predict impact sensitivities of various types of explosive molecules. Two hundreds and thirty four explosive molecules have been taken from a single database, and thirty nine molecular descriptors were computed for each explosive molecule. Optimization of NN architecture has been carried out by examining seven different sets of molecular descriptors and varying the number of hidden neurons. For the optimized NN architecture, we have utilized 17 molecular descriptors which were composed of compositional and topological descriptors in an input layer, and 2 hidden neurons in a hidden layer.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.113-117
• /
• 1998

We propose a way to show the inherent learning complexity for the multilayer perceptron. We display the solution space and the error surfaces on the input space of a single neuron with two inputs. The evolution of its weights will follow one of the two error surfaces. We observe that when we use the back-propagation(BP) learning algorithm (1), the wight cam not jump to the lower error surface due to the implicit continuity constraint on the changes of weight. The self-relaxation approach is to explicity find out the best combination of all neurons' two error surfaces. The time complexity of training a multilayer perceptron by self-relaxationis exponential to the number of neurons.