• Communications for Statistical Applications and Methods
• /
• 제6권2호
• /
• pp.511-522
• /
• 1999

The paper considers the problem of robustness in hierarchical bayesian models. In specific we address Bayesian robustness in the estimation of normal means. We provide the ranges of the posterior means under $\varepsilon$-contamination class as well as the density ratio class of priors. For the class of priors that are uniform over a specified interval we investigate the sensitivity as to the choice of the intervals. The methods are illustrated using the famous baseball data of Efron and Morris(1975).

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.51.1-51
• /
• 2001

Teleoperation is a rapidly progressing field promising to have significant application in space, mining, medical, and other areas. Unfortunately, significant communication delays are expected in teleoperation. With this problem, the two major issues in teleoperation are stability robustness and transparency performance. Recent work in bilateral teleoperation in the two-channel showed that for any two-channel control architecture, stability robustness is enhanced if the feedforward control parameters are lowered. In this paper, we analyze the stability and performance robustness of the three channels control architectures using the passivity-based Llewellyn´s absolute stability criterion as wall as the minimum values and Z-width´s of the operator and environment transmitted impedances. And the stability and the performance robustness of two and three channels control architectures are quantitatively compared.

• 한국자동차공학회논문집
• /
• 제19권6호
• /
• pp.133-139
• /
• 2011

Maintaining adequate sealing in engine cylinder head is a crucial factor in engine design. Failure of engine operations occurs mainly owing to the leaking by decreased sealing pressure. Reliability-robustness concept is applied to the engine cylinder head system. Deterministic way to obtain engineering solution in CAE industry may not consider the effects of noises and disturbances experienced during operation. However, analytical reliability-robustness concept may make possible to reduce the sensitivity of system with noise factors. Influences of design factors including noise factors would be predicted in analytical way. Optimized design may be obtained by shrinking variability and shifting to design target. Three-dimensional finite element analyses have been performed to apply analytical reliability-robustness concept.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1998년도 제13차 학술회의논문집
• /
• pp.455-460
• /
• 1998

We propose robust MILP model for scheduling and design of multiproduct batch processes in this paper. Recent stochastic modeling approaches considering uncertainty have mainly focused on maximization of expected NPV. Robust model concept is applied to generate solution spectrum in which we can select the best solution based on tradeoff between robustness measure and expected NPV. Robustness measure is represented as penalty term in the objective function, which is Upper Partial Mean of NPV. We can quantify solution robustness by this penalty term and maintain model as MILP form to be computationally efficient. An example illustrates the effectiveness of the proposed model. In many cases sufficient robustness can be guaranteed through a little reduction of expected NPV.

• 방송공학회논문지
• /
• 제26권7호
• /
• pp.862-867
• /
• 2021

Although Deep Neural Networks (DNNs) have shown remarkable performance in various artificial intelligence fields, it is well known that DNNs are vulnerable to adversarial attacks. Since adversarial attacks are implemented by adding perturbations onto benign examples, increasing the sparsity of DNNs minimizes the propagation of errors to high-level layers. In this paper, unlike the traditional pruning scheme removing low magnitude weights, we eliminate high magnitude weights that are usually considered high absolute values, named 'reverse pruning' to ensure robustness. By conducting both theoretical and experimental analyses, we observe that reverse pruning ensures the robustness of DNNs. Experimental results show that our reverse pruning outperforms previous work with 29.01% in Top-1 accuracy on perturbed CIFAR-10. However, reverse pruning does not guarantee benign samples. To relax this problem, we further conducted experiments by adding a regularization term for the high magnitude weights. With adding the regularization term, we also applied conventional pruning to ensure the robustness of DNNs.

• Asia pacific journal of information systems
• /
• 제31권3호
• /
• pp.335-357
• /
• 2021

As reproduction of images can be done with ease, copy detection has increasingly become important. In the duplication process, image modifications are likely to occur and some alterations are deliberate and can be viewed as attacks. A wide range of copy detection techniques has been proposed. In our study, content-based copy detection, which basically applies DCT-based features for images, namely, pixel values, edges, texture information and frequency-domain component distribution, is employed. Experiments are carried out to evaluate robustness and sensitivity of DCT-based features from attacks. As different types of DCT-based features hold different pieces of information, how features and attacks are related can be shown in their robustness and sensitivity. Rather than searching for proper features, use of robustness and sensitivity is proposed here to realize how the attacked features have changed when an image attack occurs. The experiments show that, out of ten attacks, the neural networks are able to detect seven attacks namely, Gaussian noise, S&P noise, Gamma correction (high), blurring, resizing (big), compression and rotation with mostly related to their sensitive features.

• 한국컴퓨터정보학회논문지
• /
• 제24권10호
• /
• pp.205-213
• /
• 2019

네트워크 강건성은 네트워크의 가장 중요한 요구특성 중 하나이다. 전술 통신 네트워크에서 강건성은 공격 단계를 지속적으로 유지하는 핵심 기능이다. TICN이라고 하는 육군의 전술 통신 네트워크는 격자형과 트리형 네트워크 토폴로지가 혼합되어 있으며, 트리형 구조는 네트워크 강건성에서 다소 약하다. 따라서 본 논문에서는 그래프 이론을 적용하여 현재 전술 통신 네트워크의 취약점인 단절점과 브리지를 탐색하고 이 부분을 개선하기 위해 기존 네트워크에 링크를 추가하여 네트워크를 생성한다. 이후 SNA 기반 네트워크 강건성 평가를 위해 핵심 노드를 단절하여 각 네트워크의 지표 변화를 관찰한다. 실험 결과, 보완된 네트워크의 강건성이 기존 운용개념 하의 네트워크 구조보다 개선되었음을 보여준다.

• 한국정보과학회논문지:시스템및이론
• /
• 제37권3호
• /
• pp.138-146
• /
• 2010

생체네트워크는 여러 종류의 환경 변화에 매우 강건하다고 알려져 있지만 그 메커니즘은 아직 밝혀지지 않고 있다. 본 논문에서는 랜덤 네트워크에 비해 생체네트워크에 되먹임 루프가 매우 많이 존재한다는 구조적 특징을 발견하고 그것이 네트워크의 강건성에 어떤 영향을 미치는지를 살펴보았다. 이를 위해 불리언 네트워크 모델을 이용하여 네트워크 강건성을 적절하게 측정하는 방법을 정의하고 많은 불리언 네트워크에 대해서 시뮬레이션하였다. 그 결과, 불리언 네트워크에서 되먹임 루프의 개수가 증가하면 고정점 끌개의 개수는 거의 변화가 없지만 유한순환 끌개의 개수는 크게 줄어든다는 사실을 밝혔다. 또한, 되먹임 루프의 개수가 증가함에 따라 고정점 끌개로 수렴하는 거대한 끌개 영역이 생성됨을 보였다. 이러한 사실들은 매우 많은 수의 되먹임 루프가 네트워크의 강건성을 높이는 데 중요한 요인임을 설명한다.

• 대한기계학회논문집A
• /
• 제27권7호
• /
• pp.1234-1242
• /
• 2003

In this paper we present a simple and efficient robust optimal design formulation for MEMS structures and its application to a resonant-type micro probe. The basic idea is to use the gradient index (GI) to improve robustness of the objective and constraint functions. In the robust optimal design procedure, a deterministic optimization for performance of MEMS structures is followed by design sensitivity analysis with respect to uncertainties such as fabrication errors and change of operating conditions. During the process of deterministic optimization and sensitivity analysis, dominant performance and uncertain variables are identified to define GI. The GI is incorporated as a term of objective and constraint functions in the robust optimal design formulation to make both performance and robustness improved. While most previous approaches for robust optimal design require statistical information on design variations, the proposed GI based method needs no such information and therefore is cost-effective and easily applicable to early design stages. For the micro probe example, robust optimums are obtained to satisfy the targets for the measurement sensitivity and they are compared in terms of robustness and production yield with the deterministic optimums through the Monte Carlo simulation. This method, although shown for MEMS structures, may as well be easily applied to conventional mechanical structures where information on uncertainties is lacking but robustness is highly important.

• 전기학회논문지
• /
• 제60권10호
• /
• pp.1924-1934
• /
• 2011

This paper is concerned with the robustness evaluations of the guidance controller for a bimodal tram which is being developed by the Korea Railroad Research Institute (KRRI). The bimodal tram is an all-wheel steered multiple-articulated vehicle as a new kind of transportation vehicle. This vehicle has to be equipped with an automatic guidance system. In [1], such a controller has been recently proposed. However, since the performance is affected by weight change of the vehicle due to number of the passenger, model parameter uncertainties depending on the state of friction and the elasticity of the tire, and a typhoon, the controller designed must be examined with these conditions. As expected, because the vehicle dynamics is highly nonlinear, for the sake of investigating the robustness of the controller we compose two simulation ways based on the vehicle models which are implemented by the ADAMS and the MATLAB/LabVIEW toolboxes. Different uncertainties and a typhoon disturbance have been considered for the simulation conditions. Simulation results are shown.