• 문화기술의 융합
• /
• 제10권1호
• /
• pp.435-441
• /
• 2024

수집된 대량의 데이터셋이 딥러닝 학습데이터로 사용되기 위해서는 주민번호, 질병 정보등과 같이 민감한 개인정보는 해커에게 노출되지 않도록 값을 변경하거나 암호화해야 하고 구축된 딥러닝 모델의 구조와 일치 하도록 데이터를 재구성 해주어야 한다. 현재, 이러한 작업은 전문가에 의해 수동으로 이루어지기 때문에, 시간과 비용이 많이 소요 된다. 이러한 문제점을 해결하기 위해, 본 논문에서는 딥러닝 과정에서 개인정보 보호를 위한 데이터 처리 작업을 자동으로 수행할 수 있는 기법을 제안한다. 제안된 기법에서는 데이터 일반화에 기반한 개인정보 보호 작업을 수행하고 원형큐를 사용하여 데이터 재구성 작업을 수행한다. 제안된 기법의 타당성을 검증하기 위해, C언어를 사용하여 직접 구현하였다. 검증 결과, 데이터 일반화가 정상적으로 수행되고 딥러닝 모델에 맞는 데이터 재구성이 제대로 수행됨을 확인 할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.230-235
• /
• 1992

We propose a method of generating data to train a neural network controller. The data can be prepared directly by an iterative learning technique which repeatedly adjusts the control input to improve the tracking quality of the desired trajectory. Instead of storing control input data in memory as in iterative learning control, the neural network stores the mapping between the control input and the desired output. We apply this concept to the trajectory control of a two link robot manipulator with a feedforward neural network controller and a feedback linear controller. Simulation results show good generalization of the neural network controller.

• International Journal of Control, Automation, and Systems
• /
• 제1권3호
• /
• pp.289-300
• /
• 2003

In this paper, we introduce an identification method in Fuzzy Relation-based Fuzzy Neural Networks (FRFNN) through a hybrid identification algorithm. The proposed FRFNN modeling implement system structure and parameter identification in the efficient form of "If...., then... " statements, and exploit the theory of system optimization and fuzzy rules. The FRFNN modeling and identification environment realizes parameter identification through a synergistic usage of genetic optimization and complex search method. The hybrid identification algorithm is carried out by combining both genetic optimization and the improved complex method in order to guarantee both global optimization and local convergence. An aggregate objective function with a weighting factor is introduced to achieve a sound balance between approximation and generalization of the model. The proposed model is experimented with using two nonlinear data. The obtained experimental results reveal that the proposed networks exhibit high accuracy and generalization capabilities in comparison to other models.er models.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.1557-1560
• /
• 1997

In this paper, we have built an experimental magnetic levitation system for a possible use of control education. We have give a mathermatical model of the nonlinear system and have shown the stability region of the linearized system when it is controlled by a PD controller. We also proposed a neural network control system which uses a neural network as a feedforward controller thgether with a conventional feedback PF controller. We have generated a desired output trajectory, which was designed for the benefit of the generalization of the neural network controller, and trained the desired output trajectory, which was desigend for the benefit of the generalization of the neural netowrk controller, and trained a neural network controller with the data of the actual input and the output of the system obtained by applying the desired output trajectroy. A good tracking performance was observed for both the desired trajectiories used and not used for the neural network training.

• 한국측량학회:학술대회논문집
• /
• 한국측량학회 2006년도 춘계학술발표회 논문집
• /
• pp.405-410
• /
• 2006

The digital map version 2.0 is national base map which is used for frame work data, paper map making as well as geographic information system. National Geographic Institude has been research to make small scale digital map by using large scale digital map. NGI made from 25 1/5,000 digital maps to one 1/25,000 digital map ver 2.0 with map generalization system in 2003. However, they could not make 1/10,000 and l/50,000 digital map version 2.0 because of There is no portrayal on the scale 1/10,000 and 1/50,000 digital map in the existing regulations. therefore. We create the specification of the digital map on scale in order to make small scale digital map version 2.0 such as 1/10,000 and 1/50,000 scale.

• Smart Structures and Systems
• /
• 제15권4호
• /
• pp.1103-1120
• /
• 2015

A self-sensing magnetorheological (MR) damper with embedded piezoelectric force sensor has recently been devised to facilitate real-time close-looped control of structural vibration in a simple and reliable manner. The development and characterization of the self-sensing MR damper are presented based on experimental work, which demonstrates its reliable force sensing and controllable damping capabilities. With the use of experimental data acquired under harmonic loading, a nonparametric dynamic model is formulated to portray the nonlinear behaviors of the self-sensing MR damper based on NARX modeling and neural network techniques. The Bayesian regularization is adopted in the network training procedure to eschew overfitting problem and enhance generalization. Verification results indicate that the developed NARX network model accurately describes the forward dynamics of the self-sensing MR damper and has superior prediction performance and generalization capability over a Bouc-Wen parametric model.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.141-144
• /
• 2008

Many studies have been conducted on extracting buildings from ALS(Airborne Laser Scanning) data. After segmentation or classification of building points, additional steps such as generalization is required to get straight boundary lines that better approximate the real ones. In much research, orthogonal constraints are used to improve accuracies and qualities. All the lines of the building boundaries are assumed to be either parallel or perpendicular mutually. However, this assumption is not valid in many cases and more complex shapes of buildings have been increased. A new algorithm is presented that is applicable to various complex buildings. It consists of three steps of boundary tracing, grouping, and regularization. The performance of our approach was evaluated by applying the algorithm to some buildings and the results showed that our proposed method has good potential for extracting building boundaries of various shapes.

• 한국해양공학회지
• /
• 제10권1호
• /
• pp.65-74
• /
• 1996

Up to now, the existing crack growth modelling has used a mathematical approximation but an assumed function have a great influence on this method. Especially, crack growth behavior that shows very strong nonlinearity needed complicated function which has difficulty in setting parameter of it. The main characteristics of neural network modelling to engineering field are simple calculations and absence of assumed function. In this paper, after discussing learning and generalization of neural networks, we performed crack growth modelling on the basis of above learning algorithms. J'-da/dt relation predicted by neural networks shows that test condition with unlearned data is simulated well within estimated mean error(5%).

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.484-484
• /
• 2000

The problem of determining the proper size of an neural network is recognized to be crucial, especially for its practical implications in such important issues as learning and generalization. Unfortunately, it usually is not obvious what size is best: a system that is too snail will not be able to learn the data while one that is just big enough may learn the slowly and be very sensitive to initial conditions and learning parameters. One popular technique is commonly known as pruning and consists of training a larger than necessary network and then removing unnecessary weights/nodes. In this paper, a new pruning method is developed, based on the penalty-term methods. This method makes the neural network good for the generalization and reduces the retraining time after pruning weights/nodes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권5호
• /
• pp.877-895
• /
• 2010

The state-of-the-art techniques of storing XML data, modeled as an XML tree, are node-based in the sense that they are centered around XML node labeling and the storage unit is an XML node. In this paper, we propose a generalization of such techniques so that the storage unit is an XML subtree that consists of one or more nodes. Despite several advantages with such generalization, a major problem would be inefficiency in XPath processing where the stored subtrees are to be parsed on the fly in order for the nodes inside them to be accessed. We solve this problem, proposing a technique whereby no parsing of the subtrees involved in XPath processing is needed at all unless they contain the nodes of the final query result. We prove that the correctness of XPath processing is guaranteed with our technique. Through implementation and experiments, we also show that the overhead of our technique is acceptable.