• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.151-162
• /
• 1994

A new neural network structure called Self-organized Distributed Networks (SODN) is proposed for developing the neural network-based multidimensional system models. The learning with the proposed networks is fast and precise. Such properties are caused from the local learning mechanism. The structure of the networks is combination of dual networks such as self-organized networks and multilayered local networks. Each local networks learns only data in a sub-region. Large number of memory requirements and low generalization capability for the untrained region, which are drawbacks of conventional local network learning, are overcomed in the proposed networks. The simulation results of the proposed networks show better performance than the standard multilayer neural networks and the Radial Basis function(RBF) networks.

• Journal of Educational Research in Mathematics
• /
• v.15 no.2
• /
• pp.215-232
• /
• 2005

Computer algebra system (CAS) calculators are becoming increasingly common in schools and universities. While Hey offer quite sophisticated mathematical capability to teachers and students, it is not clear at present how they may best be employed. In particular their integration into students' learning and problem-solving remains an issue. In this paper we address this issue through the lens of a study that considered the introduction of the TI-89 CAS calculator to students about to enter university. We describe a number of different aspects of the partnership they formed with the calculator as they began the process of instrumentation of the CAS in their learning.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.2
• /
• pp.56-64
• /
• 1998

Multi-layered perceptrons that are a nonlinear neural network model, have been widely used for various applications mainly thanks to good function approximation capability for nonlinear fuctions. However, for digital hardware implementation of the multi-layere perceptrons, the quantization scheme using "look-up tables (LUTs)" is commonly employed to handle nonlinear signmoid activation functions in the neworks, and thus requires large amount of storage to prevent unacceptable quantization errors. This paper is concerned with a new effective methodology for digital hardware implementation of multi-layered perceptrons, and proposes a "piecewise affine approximation" method in which input domain is divided into (small number of) sub-intervals and nonlinear sigmoid function is linearly approximated within each sub-interval. Using the proposed method, we develop an expression and an error backpropagation type learning algorithm for a multi-layered perceptron, and compare the performance with the quantization method through Monte Carlo simulations on XOR problems. Simulation results show that, in terms of learning convergece, the proposed method with a small number of sub-intervals significantly outperforms the quantization method with a very large storage requirement. We expect from these results that the proposed method can be utilized in digital system implementation to significantly reduce the storage requirement, quantization error, and learning time of the quantization method.quantization method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.5
• /
• pp.1038-1043
• /
• 2009

Recently, such education is required as the one to nurture the capability for higher order thinking such as creativity and problem-solving ability on the eduction field of not only elementary and middle school but also colleges. Thus, this study is aimed to discuss the computer programming education to enhance higher order thinking in conducting computer programming education for college students. To this end, this study is intended to suggest the computer programming education of based project learning based on preceeding studies and literature reviews. Through this research, the researcher expects that this study forms the small foundation for offering computer programing education covering the computer and engineering education across the board.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.477-484
• /
• 2001

This paper proposes a hybrid learning algorithm combined momentum and kernel-adatron for improving the performance of regression support vector machine. The momentum is utilized for high-speed convergence by restraining the oscillation in the process of converging to the optimal solution, and the kernel-adatron algorithm is also utilized for the capability by working in nonlinear feature spaces and the simple implementation. The proposed algorithm has been applied to the 1-dimension and 2-dimension nonlinear function regression problems. The simulation results show that the proposed algorithm has better the learning speed and performance of the regression, in comparison with those quadratic programming and kernel-adatron algorithm.

• Journal of the Korean School Mathematics Society
• /
• v.23 no.2
• /
• pp.223-237
• /
• 2020

Artificial intelligence technology, which represents the era of the 4th Industrial Revolution, is now deeply involved in our lives, and future education places great emphasis on building students' capabilities for the principles and uses of artificial intelligence. Therefore, the purpose of this study is to develop the contents of AI related education in mathematics, which the relationship is closely connected to each other. To this end, I propose establishing two novel AI-related contents in mathematics education. One subject is related to learning the principle of machine learning based on mathematics foundation. In addition, I draw the core math contents dealt in following subject called 'Basic Mathematics for AI and Data Science.'

• Korean Journal of Computational Design and Engineering
• /
• v.5 no.2
• /
• pp.127-135
• /
• 2000

Emergent computing paradigms such as genetic algorithms have found increased use in problems in engineering design. These computational tools have been shown to be applicable in the solution of generically difficult design optimization problems characterized by nonconvexities in the design space and the presence of discrete and integer design variables. Another aspect of these computational paradigms that have been lumped under the bread subject category of soft computing, is the domain of artificial intelligence, knowledge-based expert system, and machine learning. The paper explores a machine learning paradigm referred to as teaming classifier systems to construct the high-quality global function approximations between the design variables and a response function for subsequent use in design optimization. A classifier system is a machine teaming system which learns syntactically simple string rules, called classifiers for guiding the system's performance in an arbitrary environment. The capability of a learning classifier system facilitates the adaptive selection of the optimal number of training data according to the noise and multimodality in the design space of interest. The present study used the polynomial based response surface as global function approximation tools and showed its effectiveness in the improvement on the approximation performance.

• Nuclear Engineering and Technology
• /
• v.52 no.2
• /
• pp.287-295
• /
• 2020

Group method of data handling (GMDH) is considered one of the earliest deep learning methods. Deep learning gained additional interest in today's applications due to its capability to handle complex and high dimensional problems. In this study, multi-layer GMDH networks are used to perform uncertainty quantification (UQ) and sensitivity analysis (SA) of nuclear reactor simulations. GMDH is utilized as a surrogate/metamodel to replace high fidelity computer models with cheap-to-evaluate surrogate models, which facilitate UQ and SA tasks (e.g. variance decomposition, uncertainty propagation, etc.). GMDH performance is validated through two UQ applications in reactor simulations: (1) low dimensional input space (two-phase flow in a reactor channel), and (2) high dimensional space (8-group homogenized cross-sections). In both applications, GMDH networks show very good performance with small mean absolute and squared errors as well as high accuracy in capturing the target variance. GMDH is utilized afterward to perform UQ tasks such as variance decomposition through Sobol indices, and GMDH-based uncertainty propagation with large number of samples. GMDH performance is also compared to other surrogates including Gaussian processes and polynomial chaos expansions. The comparison shows that GMDH has competitive performance with the other methods for the low dimensional problem, and reliable performance for the high dimensional problem.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.33 no.4
• /
• pp.290-297
• /
• 1997

Underwater robotic vehicles(URVs) are used for various work assignments such as pipe-lining, inspection, data collection, drill support, hydrography mapping, construction, maintenance and repairing of undersea equipment, etc. As the use of such vehicles increases the development of vehicles having greater autonomy becomes highly desirable. The vehicle control system is one of the most critic vehicle subsystems to increase autonomy of the vehicle. The vehicle dynamics is nonlinear and time-varying. Hydrodynamic coefficients are often difficult to accurately estimate. It was also observed by experiments that the effect of electrically powered thruster dynamics on the vehicle become significant at low speed or stationkeeping. The conventional linear controller with fixed gains based on the simplified vehicle dynamics, such as PID, may not be able to handle these properties and result in poor performance. Therefore, it is desirable to have a control system with the capability of learning and adapting to the changes in the vehicle dynamics and operating parameters and providing desired performance. This paper presents an adaptive and learning control system which estimates a new set of parameters defined as combinations of unknown bounded constants of system parameter matrices, rather than system parameters. The control system is described with the proof of stability and the effect of unmodeled thruster dynamics on a single thruster vehicle system is also investigated.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.25-28
• /
• 2006

사용자 의도 파악 (intention reading) 기술은 스마트 홈과 같은 복잡한 유비쿼터스(ubiquitous) 환경에서 사용자에게 보다 편리하고 개인화된(personalized) 서비스 제공이 가능하도록 해준다. 또한 학습 기능(learning capability)은 지식 발견(knowledge discovery)의 관점에서 의도 파악 기술의 핵심 요소 기술의 하나로 자리 매김 하고 있다. 본 논문에서는 스마트 홈 환경에서 제공 가능한 개인화된 서버스(personalized service) 중의 하나로, 개인화된 미디어 제어 방법에 대한 내용을 다룬다. 특히, 이러한 사람의 행동 패턴과 같은 데이터는 패턴 분류의 관점에서 구분해야 할 클래스(class)에 비해 입력 정보가 불충분할 경우가 많으므로 비일관적인(inconsistent) 데이터가 많으므로, 퍼지 논리(fuzzy logic)와 확률(probability)의 개념을 효과적으로 병행해야 의미 있는 지식을 추출해 낼 수 있다. 이를 위하여 반복 퍼지 지도 클러스터링 (IFCS; Iterative Fuzzy Clustering with Supervision) 알고리즘에 기반하여 주어진 데이터 패턴으로부터 확률적 퍼지 룰(probabilistic fuzzy rule)을 얻어 내는 방법에 대해 설명한다. 또한 이를 포함하는 학습 제어 시스템을 통해 개인화된 미디어 서비스를 추천해 줄 수 있는 방법에 대해서 설명하도록 한다.