• 생물정신의학
• /
• 제14권4호
• /
• pp.221-231
• /
• 2007

Recent researches have shown that important cellular-based autoprotective mechanisms are mediated by heat-shock proteins(HSPs), also called 'molecular chaperones'. HSPs as molecular chaperones are the primary cellular defense mechanism against damage to the proteome, initiating refolding of denatured proteins and regulating degradation after severe protein damage. HSPs also modulate multiple events within apoptotic pathways to help sustain cell survival following damaging stimuli. HSPs are induced by almost every type of stresses including physical and psychological stresses. Our nervous system in the brain are more vulnerable to stress and damage than any other tissues due to HSPs insufficiency. The normal function of HSPs is a key factor for endogenous stress adaptation of neural tissues. HSPs play an important role in the process of neurodevelopment, neurodegeneration, and neuroendocrine regulation. The altered function of HSPs would be associated with the development of several neuropsychiatric disorders. Therefore, an understanding of HSPs activities could help to improve autoprotective mechanism of our neural system. This paper will review the literature related to the significance of HSPs in neuropsychiatric field.

• 조명전기설비학회논문지
• /
• 제20권1호
• /
• pp.57-64
• /
• 2006

본 논문에서는 유도전동기의 벡터제어를 위한 ANN 센서리스 제어와 속도제어를 위한 AFLC를 제안하였다. AFLC 설계는 적응 메카니즘을 통해 퍼지 룰 베이스의 수정자를 갱신하여 실행할 수 있고 유도 전동기의 속도 추정을 위한 ANN 센서리스 제어는 BPA를 통해 수행하였다. 유도전동기의 지령속도와 실제속도는 BPA를 통해 그 오차를 줄일 수 있고, 이러한 알고리즘은 다른 전동기 드라이브에 적용이 용이하다. 본 논문에서 제시한 AFLC 및 ANN 제어의 응답특성을 분석하고 그 결과를 제시한다.

• 한국콘텐츠학회논문지
• /
• 제1권1호
• /
• pp.8-14
• /
• 2001

본 논문에서는 의료영상에서 특정 장기를 추출하여 질환 부위를 인식하는 알고리즘을 제안한다. 의료영상이 추출되어진 장기 부위에서 질환을 인식하기 위하여 단일 신경회로망을 이용하면 신경회로망의 학습 능력과 일반화 능력이 한정적이므로 성능개선에 많은 문제가 있다. 따라서 추출된 장기로부터 질환부위를 인식하는 것은 신경회로망을 복합적인 방법, 즉 RBF (Radial Basis Function), BP (Back Propagation)로 구성하여 단일 신경회로망의 단점을 극복하였다. 본 논문에서 제안하는 알고리즘은 입력 의료영상의 다양한 형태 변화에 적응력이 뛰어남을 실험결과로 알 수 있었다. 그리고, 전체 알고리즘의 수행시간이 장기추출 알고리즘을 포함하여 일반적으로 10초 이내에 수행됨을 실험 결과 알 수 있었다. 제안된 알고리즘은 실시간으로 의료영상의 질환부위를 인식하여 판별 자동화를 통해 원격의료에 사용 되어 질 수 있다.

• 한국전문물리치료학회지
• /
• 제2권2호
• /
• pp.73-84
• /
• 1995

Spasticity has been defined as a motor disorder characterised by a velocity-dependent increase in tonic stretch reflexes with exaggerated tendon jerks resulting in hyperexcitability of the stretch reflexes as one component of the upper motor neuron syndrome. Weakness and loss of dexterity, however, are considered to be more disabling to the patient than changes in muscle tone. The discussion includes the important role that alterations in the physiology of motor units, notably changes in firing rates and muscle fiber atrophy, play in the manifestation of muscle weakness. This paper considers both the neural and mechanical components of spasticity and discusses, in terms of clinical intervention, the implications arising from recent research. Investigations suggest that the resistance to passive movement in individuals with spasticity is due not only to neural mechanisms but also to changes in mechanical properties of muscle. The emphasis is on training the individual to gain control over the muscles required for different tasks, and on preventing secondary and adaptive soft tissue changes and ineffective adaptive motor behaviours.

• 한국지능시스템학회논문지
• /
• 제6권2호
• /
• pp.52-56
• /
• 1996

신경망을 학습하는데 있어서, 망의 학습속도는 학습율에 의해 크게 좌우된다. 그러나 대부분의 정적인 학습율 선택 방법들은 몇몇 결정적인 방법들을 제외하곤 경험적인 방식에 의존해 왔다. 경험적인 방식을 사용하여 좋은 학습율을 찾아내는 것은 배우 지류하고 어려운 일이다. 또한 결정적인 방법들은 학습율의 질을 보장하지는 못한다. 본 논문에서 우리는 새로운 학습율 선택 방법을 제안한다. 우리의 방법은 진화 프로그래밍기법을 사용하여 통계적인 방식으로 접근함으로써 좋은 학습율을 찾을 수 있다. 모의 실험을 통하여 우리의 방식이 경험적인 방식들이나 결정적인 방식보다 우수함을 보였다.

• 한국IT서비스학회지
• /
• 제20권1호
• /
• pp.81-92
• /
• 2021

As the degree of economic development of society increases, the maintenance issues on the existing social overhead capital becomes essential. Accordingly, the adaptation of the concept of Level of service in highway maintenance is indispensable. It is also crucial to manage and perform the service level such as road assets to provide universal services to users. In this regards, the purpose of this study is to improve the maintenance service rating model and to focus on the assessment items and weights among the improvements. Particularly, in determining weights, an Analytic Hierarchy Process (AHP) is performed based on the survey response results. After then, this study conducts unsupervised neural network models such as Self-Organizing Map (SOM) and Davies-Bouldin (DB) Index to divide proper sub-groups and determine priorities. This paper identifies similar cases by grouping the results of the responses based on the similarity of the survey responses. This can effectively support decision making in general situations where many evaluation factors need to be considered at once, resulting in reasonable policy decisions. It is the process of using advanced technology to find optimized management methods for maintenance.

• 한국군사과학기술학회지
• /
• 제26권1호
• /
• pp.44-53
• /
• 2023

Adversarial attacks have received great attentions for their capacity to distract state-of-the-art neural networks by modifying objects in physical domain. Patch-based attack especially have got much attention for its optimization effectiveness and feasible adaptation to any objects to attack neural network-based object detectors. However, despite their strong attack performance, generated patches are strongly perceptible for humans, violating the fundamental assumption of adversarial examples. In this paper, we propose a camouflaged adversarial patch optimization method using military camouflage assessment metrics for naturalistic patch attacks. We also investigate camouflaged attack loss functions, applications of various camouflaged patches on army tank images, and validate the proposed approach with extensive experiments attacking Yolov5 detection model. Our methods produce more natural and realistic looking camouflaged patches while achieving competitive performance.

• 대한임베디드공학회논문지
• /
• 제18권5호
• /
• pp.217-223
• /
• 2023

In this paper, we propose a new model for the conditional generation of music, considering key and rhythm, fundamental elements of music. MIDI sheet music is converted into a WAV format, which is then transformed into a Mel Spectrogram using the Short-Time Fourier Transform (STFT). Using this information, key and rhythm details are classified by passing through two Convolutional Neural Networks (CNNs), and this information is again fed into the Music Transformer. The key and rhythm details are combined by differentially multiplying the weights and the embedding vectors of the MIDI events. Several experiments are conducted, including a process for determining the optimal weights. This research represents a new effort to integrate essential elements into music generation and explains the detailed structure and operating principles of the model, verifying its effects and potentials through experiments. In this study, the accuracy for rhythm classification reached 94.7%, the accuracy for key classification reached 92.1%, and the Negative Likelihood based on the weights of the embedding vector resulted in 3.01.

• 한국지능시스템학회논문지
• /
• 제11권5호
• /
• pp.426-432
• /
• 2001

본 논문은 진화형 하드웨어를 이용하여 생물의 정보처리 시스템인 셀룰라 오토마타 신경망의 구현에 관한 연구이다. 셀룰라 오토마타 신경망은 진화 및 발생을 기반으로 한 신경망 모델이다. 진화는 다양성을 주요 근원을 제공하는 돌연변이 및 재 조합 비율에 의하여 비결정론이며, 발생은 결정론 적이며 지역적인 무리현상을 따른다. 셀룰라 오토마타 신경망은 셀룰라 오토마타에 의해 신경망 내부의 각 셀의 상태를 발생시키고, 초기 셀을 유전자 알고리즘의 개체로 간주하여 초기 셀이 진화 알고리즘을 통해 진화함으로써 신경망이 진화하는 시스템이다. 본 논문은 이 시스템을 진화형 하드웨어 이용하여 하드웨어로 구현하였다. 진화형 하드웨어는 진화 알고리즘과 재구성하드웨어의 결합체이다. 즉, 재구성 하드웨어의 구성에 필요한 bit를 유전자 알고리즘의 개체로 간주한 것이다. 진화 알고리즘을 수행하기 위해 유전자 알고리즘 프로세서를 설계하였으며, 셀룰라 오토마타 신경망이 유전자 알고리즘의 개체와 셀룰라 오토마타 룰에 의해 자동적으로 신경망을 생성하기 위해 신경망을 이루는 셀들로 설계하였다. 제안된 시스템의 효율성을 검증하기 위해 Exclusive-OR 문제에 적용하였다.

• Nuclear Engineering and Technology
• /
• 제51권3호
• /
• pp.807-817
• /
• 2019

Airborne Gamma Ray Spectrometry (AGRS) with its important applications such as gathering radiation information of ground surface, geochemistry measuring of the abundance of Potassium, Thorium and Uranium in outer earth layer, environmental and nuclear site surveillance has a key role in the field of nuclear science and human life. The Broyden-Fletcher-Goldfarb-Shanno (BFGS), with its advanced numerical unconstrained nonlinear optimization in collaboration with Artificial Neural Networks (ANNs) provides a noteworthy opportunity for modern AGRS. In this study a new AGRS system empowered by ANN-BFGS has been proposed and evaluated on available empirical AGRS data. To that effect different architectures of adaptive ANN-BFGS were implemented for a sort of published experimental AGRS outputs. The selected approach among of various training methods, with its low iteration cost and nondiagonal scaling allocation is a new powerful algorithm for AGRS data due to its inherent stochastic properties. Experiments were performed by different architectures and trainings, the selected scheme achieved the smallest number of epochs, the minimum Mean Square Error (MSE) and the maximum performance in compare with different types of optimization strategies and algorithms. The proposed method is capable to be implemented on a cost effective and minimum electronic equipment to present its real-time process, which will let it to be used on board a light Unmanned Aerial Vehicle (UAV). The advanced adaptation properties and models of neural network, the training of stochastic process and its implementation on DSP outstands an affordable, reliable and low cost AGRS design. The main outcome of the study shows this method increases the quality of curvature information of AGRS data while cost of the algorithm is reduced in each iteration so the proposed ANN-BFGS is a trustworthy appropriate model for Gamma-ray data reconstruction and analysis based on advanced novel artificial intelligence systems.