• 제목/요약/키워드: Trade-Off Theory

검색결과 77건 처리시간 0.025초

Human Detection 을 위한 Bayesian Logistic Regression (Bayesian Logistic Regression for Human Detection)

  • ;;이칠우
    • 한국HCI학회:학술대회논문집
    • /
    • 한국HCI학회 2008년도 학술대회 1부
    • /
    • pp.569-572
    • /
    • 2008
  • The possibility to extent the solution in human detection problem for plug-in on vision-based Human Computer Interaction domain is very attractive, since the successful of the machine leaning theory and computer vision marriage. Bayesian logistic regression is a powerful classifier performing sparseness and high accuracy. The difficulties of finding people in an image will be conquered by implementing this Bavesian model as classifier. The comparison with other massive classifier e.g. SVM and RVM will introduce acceptance of this method for human detection problem. Our experimental results show the good performance of Bavesian logistic regression in human detection problem, both in trade-off curves (ROC, DET) and real-implementation compare to SVM and RVM.

  • PDF

TRAJECTORY CONTROL OF ROBOT MANIPULATORS USING VSS THEDORY smoothing modification : SMOOTHING MODIFICATION

  • Hideki Hashimoto;Sim, Kwee-Bo;Jianxin Xu;F. Harashima
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 1987년도 한국자동제어학술회의논문집(한일합동학술편); 한국과학기술대학, 충남; 16-17 Oct. 1987
    • /
    • pp.898-904
    • /
    • 1987
  • A new control algorithm using the VSS theory is developed for accurate trajectory control of robot manipulators. This paper focuses on the implementation of VSS controller with smoothing laws in the design of effective tracking control for robotic arms. The VSS controller for multi-linkage robot arm is realized using balance condition and its simplification which possesses powerful smoothing capability to reduce or even remove undesirable chattering and meanwhile keep the robust characteristic to reject system uncertainties. The design principle of selecting different smoothing methods, which aims at achieving trade-off between smoothing and robust behaviors while considering the actual system constraints, is also given and further confirmed through experimental results.

  • PDF

A dynamic approach to manufacturing improvement from learning and decision-theoretic perspectives

  • Kim, Bowon
    • 한국경영과학회:학술대회논문집
    • /
    • 한국경영과학회 1996년도 추계학술대회발표논문집; 고려대학교, 서울; 26 Oct. 1996
    • /
    • pp.49-52
    • /
    • 1996
  • In this article, we develop a 'dynamic' approach to manufacturing improvement, based on perspectives of manufacturing learning and decision theory. First, we present an alternative definition of production system consistent with a decision-theoretic perspective: the system consists of structural, infra-structural, and decision making constructs. A primary proposition is that learning capability possessed by a manufacturing system be prerequisite for the system to improve its manufacturing performance through optimal controlling of the three constructs. To support the proposition, we elaborate on a mathematical representation of "learning" as defined in an applied setting. We show how the learning capability acts as an integrating force ameliorating the trade-off between two key manufacturing capabilities, i.e., process controllability and process flexibility.exibility.

  • PDF

반복제어기를 이용한 하드디스크의 주기적인 외란 보상 (Compensation of the Repeatable Run Out Using Repetitive Controller in HDD)

  • 신호철;박성원;박태욱;양현석;박영필
    • 한국소음진동공학회논문집
    • /
    • 제14권2호
    • /
    • pp.136-143
    • /
    • 2004
  • This paper presents three algorithms of repetitive controller for compensation of the repeatable runout in hard disk drive. The basic theory of the repetitive controller and the analysis of the disturbance in hard disk drive are introduced. The tracking controller is designed in order to design the "plug-in type" repetitive controller. Design of the repetitive controller is considered as the design of the filter, determination of gain and design of additional compensation for the various types. Specially, trade-off relationship between stability and performance is important for the design. The three kinds of "plug-in type" repetitive controllers are designed, simulated and experimented. Performance and characteristic of them are compared with the analysis of the experimental results.

반복제어기를 이용한 하드디스크의 주기적인 외란 보상 (Compensation of the Repeatable Run Out using Repetitive Controller in HDD)

  • 신호철;박성원;박태욱;양현석;박영필
    • 한국소음진동공학회:학술대회논문집
    • /
    • 한국소음진동공학회 2003년도 추계학술대회논문집
    • /
    • pp.1083-1088
    • /
    • 2003
  • This paper presents three algorithms of repetitive controller for compensation of the repeatable uncut in hard disk drive. The basic theory of the repetitive controller and the analysis of the disturbance in hard disk drive are introduced. The tracking controller is designed in order to design the "plug-in type" repetitive controller. Design of the repetitive controller is considered as the design of the filter, determination of Gain and design of additional compensation for the various types. Specially, trade-off relationship between stability and performance is important factor for the design. The thee kinds of "plug-in type" repetitive controllers are designed simulated and experimented. Performance and characteristic of them are compared by the analysis of the experimental results

  • PDF

Cautious View on Network Coding - From Theory to Practice

  • Heide, Janus;Pedersen, Morten V.;Fitzek, Frank H.P.;Larsen, Torben
    • Journal of Communications and Networks
    • /
    • 제10권4호
    • /
    • pp.403-411
    • /
    • 2008
  • Energy consumption has been mostly neglected in network coding (NC) research so far. This work investigates several different properties of NC that influence the energy consumption and thus are important when designing NC systems for battery-driven devices. Different approaches to the necessary implementation of coding operations and Galois fields arithmetic are considered and complexity expressions for coding operations are provided. We also benchmark our own mobile phone implementation on a Nokia N95 under different settings. Several NC strategies are described and compared, furthermore expressions for transmission times are developed. It is also shown that the use of NC introduces a trade off between reduction in transmission time and increase in energy consumption.

농업발전(農業發展) 이론연구(理論硏究) (I) - Fei-Ranis의 경제사적(經濟史的) 접근방법(接近方法)을 중심(中心)으로 - (An Inquiry into Agricultural Development Theory (1) - Fei-Ranis's Historical Approach and its Relevance to Less Developed World -)

  • 이호철
    • Current Research on Agriculture and Life Sciences
    • /
    • 제1권
    • /
    • pp.239-253
    • /
    • 1983
  • This study attempted to introduce Fei-Ranis's agricultural development theory and discuss its problem for the rural development of less developed world. Fei-Ranis systematized the development process of Western European economy on the ground of dualism. They divided the process into 4 stages by the concept of 'mode of operation'. Paticularly, they consider agrarian mercantilism as take-off stage and its development were achieved by the increase of trade margin and labor productivity. Especially, they thought that only agricultural revolution through the diffusion of internal exchange economy and construction of tree-star system can accomplish favorable transition to industrial capitalism. In order to promote this agricultural development, less developed world must abolish short-run agricultural policy and propel 'learning by the contact' strategy through 'tree-star system' and 'parellel development.' In reality, it was problematic that the contemporary less developed world is trying, in the course of a few decades, to imitate Western European experience with development over the last four centuries. But Fei-Ranis ignored qualitative aspects of agricultural development by tree-star system and also it is criticized that they considered agricultural development process of less developed world follows only that of Western European classical process.

  • PDF

제한된 무선 자원 환경에 적합한 주파수 자동지정 알고리즘 (A Novel Frequency Allocation Algorithm for Limited Radio Resource Environments)

  • 구본홍;채찬병;박성호;박휘성;함재현
    • 한국통신학회논문지
    • /
    • 제40권9호
    • /
    • pp.1719-1721
    • /
    • 2015
  • 주파수 자동지정 알고리즘은 빈번한 환경 변화에 따라 주파수 재지정이 요구되는 군용 통신에서 중요하게 고려되고 있다. 본 논문에서는 주파수 자동지정 문제를 그래프 이론 관점에서 해결한다. 그래프의 색칠 이론을 기반으로 한 알고리즘을 제시하여 기존의 탐욕 알고리즘 보다 사용되는 주파수 개수를 0.77배 감소시키는 성능 향상을 확인하였다. 또한, 주파수 범위와 재사용률 간의 맞교환을 통하여 원하는 성능을 얻을 수 있는 하이브리드 알고리즘을 제시하였다.

Impact of Capital Structure on Profitability: A Comparative Study of Islamic and Conventional Banks of Pakistan

  • QAYYUM, Noor ul;NOREEN, Umara
    • The Journal of Asian Finance, Economics and Business
    • /
    • 제6권4호
    • /
    • pp.65-74
    • /
    • 2019
  • This study has two main purposes; first, it examines the effect of capital structure on profitability of Islamic and conventional banks; second, it determines that whether the capital structure of Islamic and conventional banks is same or not. A sample of ten banks was taken over the period 2006-2016. Independent samples T-test was used for finding the comparison between the capital structure of Islamic and conventional banks while for assessing the impact of capital structure on profitability, regression analysis (Fixed effects model) was used. Results showed that the capital structure of both types of banks was similar except for bank size which differed significantly. Moreover, ROA was negatively correlated to the capital structure of both conventional and Islamic banks. In contrast, ROE was positively correlated to the capital structure of both conventional and Islamic banks. In addition to that, two explanatory variables were positively correlated while two were negatively correlated to EPS for both Islamic and conventional banks. This study proves the existence of prominent theories of capital structure (pecking order theory and trade-off theory) for both conventional and Islamic banks in Pakistan and also validates the economies of scale.

지구물리자료의 역산해석에 관한 개관 (Review on the inversion Analysis of Geophysical Data)

  • 김희준;정승환
    • 지구물리와물리탐사
    • /
    • 제2권2호
    • /
    • pp.112-121
    • /
    • 1999
  • This article reviews the development of geophysical inverse theory. In a series of articles published in 1967, 1968, and 1979, G. Backus and F. Gilbert a trade-off between model resolution and estimation errors in geophysical inverse problems, and gave a criterion to compromise the reciprocal relation. Although the criterion was not clear in the physical point of view, it had been extensively used in the interpretation of geophysical date in the 1970s. This was the starting point of the fruitful development of inverse theory in geophysics. A reasonable criterion to compromise the reciprocal relation was derived to solve linear problems by D. D. jackson in 1979, introducing the concept of a priori information about unknown model parameters. This Jackson's approach was extended to solve nonlinear problems on the basis o probabilistic approach to the inverse problems formulated by A. Tarantola and B. Vallete in 1982. At the end of 1980s ABIC (Akaike Bayesian Information Criterion) was introduced for selecting a more reasonable model in geophysics. Now the date inversion is regarded as the process of extracting new information from observed data, combining in with a priori information about model parameters, and constructing a more clear image of model.

  • PDF