• Communications for Statistical Applications and Methods
• /
• 제16권3호
• /
• pp.519-528
• /
• 2009

This paper deals with nonparametric estimation of discontinuous regression curve. Quite number of researches about this topic have been done. These researches are classified into two categories, the indirect approach and direct approach. The major goal of the indirect approach is to obtain good estimates of jump locations, whereas the major goal of the direct approach is to obtain overall good estimate of the regression curve. Thus it seems that two approaches are quite different in nature, so people say that the comparison of two approaches does not make much sense. Therefore, a thorough comparison of them is lacking. However, even though the main issue of the indirect approach is the estimation of jump locations, it is too obvious that we have an estimate of regression curve as the subsidiary result. The point is whether the subsidiary result of the indirect approach is as good as the main result of the direct approach. The performance of two approaches is compared through a simulation study and it turns out that the indirect approach is a very competitive tool for estimating discontinuous regression curve itself.

• Journal of Information Processing Systems
• /
• 제11권1호
• /
• pp.22-38
• /
• 2015

While Modern Standard Arabic is the formal spoken and written language of the Arab world; dialects are the major communication mode for everyday life. Therefore, identifying a speaker's dialect is critical in the Arabic-speaking world for speech processing tasks, such as automatic speech recognition or identification. In this paper, we examine two approaches that reduce the Universal Background Model (UBM) in the automatic dialect identification system across the five following Arabic Maghreb dialects: Moroccan, Tunisian, and 3 dialects of the western (Oranian), central (Algiersian), and eastern (Constantinian) regions of Algeria. We applied our approaches to the Maghreb dialect detection domain that contains a collection of 10-second utterances and we compared the performance precision gained against the dialect samples from a baseline GMM-UBM system and the ones from our own improved GMM-UBM system that uses a Reduced UBM algorithm. Our experiments show that our approaches significantly improve identification performance over purely acoustic features with an identification rate of 80.49%.

• 한국정보전자통신기술학회논문지
• /
• 제2권3호
• /
• pp.39-52
• /
• 2009

Scale Invariant Feature Transform (SIFT) is an effective algorithm in object recognition, panorama stitching, and image matching, however, due to its complexity, real time processing is difficult to achieve with software approaches. This paper proposes using a reconfigurable hardware processor with integer half kernel. The integer half kernel Gaussian reduces the Gaussian pyramid complexity in about half [] and the reconfigurable processor carries out a parallel implementation of a full search Fast SIFT algorithm. We use a low memory, fine grain single instruction stream multiple data stream (SIMD) pixel processor that is currently being developed. This implementation fully exposes the available parallelism of the SIFT algorithm process and exploits the processing and I/O capabilities of the processor which results in a system that can perform real time image and video compression. We apply this novel implementation to images and measure the effectiveness. Experimental simulation results indicate that the proposed implementation is capable of real time applications.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권6호
• /
• pp.2302-2316
• /
• 2015

We propose an efficient method of extracting targets within a region of interest in non-homogeneous infrared images by using a principal component analysis (PCA) plane and adaptive Gaussian kernel. Existing approaches for extracting targets have been limited to using only the intensity values of the pixels in a target region. However, it is difficult to extract the target regions effectively because the intensity values of the target region are mixed with the background intensity values. To overcome this problem, we propose a novel PCA based approach consisting of three steps. In the first step, we apply a PCA technique minimizing the total least-square errors of an IR image. In the second step, we generate a binary image that consists of pixels with higher values than the plane, and then calculate the second derivative of the sum of the square errors (SDSSE). In the final step, an iteration is performed until the convergence criteria is met, including the SDSSE, angle and labeling value. Therefore, a Gaussian kernel is weighted in addition to the PCA plane with the non-removed data from the previous step. Experimental results show that the proposed method achieves better segmentation performance than the existing method.

• 한국음향학회지
• /
• 제36권4호
• /
• pp.261-266
• /
• 2017

본 논문에서는 커널 모델과 장단기 기억(Long-Short Term Memory, LSTM) 신경망을 결합한 보컬 및 비보컬 분리 방식을 제안한다. 기존의 음원 분리 방식은 비보컬 음원만 있는 구간에서 음원을 오추정하여 불필요한 비보컬 음원을 출력하는 한계가 있다. 따라서 본 논문에서는 커널 모델 기반의 보컬음 분리 방식에 LSTM 신경망 기반의 보컬 구간 분류 방식을 결합하여 보컬 음원의 오추정 문제를 개선하고 분리 성능을 향상시키고자 하였다. 또한 본 논문에서는 방식간의 결합 구조에 따라 병렬 결합형 분리 알고리즘과 직렬 결합형 분리 알고리즘을 제안하였으며, 실험을 통해 제안하는 방식들이 기존의 방식에 비해 더욱 향상된 분리 성능을 보이는 것을 확인할 수 있었다.

• Communications for Statistical Applications and Methods
• /
• 제23권5호
• /
• pp.385-397
• /
• 2016

The stress-strength models have been intensively investigated in the literature in regards of estimating the reliability ${\theta}$ = P(X > Y) using parametric and nonparametric approaches under different sampling schemes when X and Y are independent random variables. In this paper, we consider the problem of estimating ${\theta}$ when (X, Y) are dependent random variables with a bivariate underlying distribution. The empirical and kernel estimates of ${\theta}$ = P(X > Y), based on bivariate ranked set sampling (BVRSS) are considered, when (X, Y) are paired dependent continuous random variables. The estimators obtained are compared to their counterpart, bivariate simple random sampling (BVSRS), via the bias and mean square error (MSE). We demonstrate that the suggested estimators based on BVRSS are more efficient than those based on BVSRS. A simulation study is conducted to gain insight into the performance of the proposed estimators. A real data example is provided to illustrate the process.

• 반도체디스플레이기술학회지
• /
• 제22권3호
• /
• pp.142-148
• /
• 2023

In this paper, we identify performance issues in executing compute kernels from PolyBench, which includes compute kernels that are the core computational units of various data-intensive workloads, such as deep learning and data-intensive applications, on Processing-in-Memory (PIM) devices. Therefore, using our in-house simulator, we measured and compared the various performance metrics of workloads based on traditional out-of-order and in-order processors with Processing-in-Memory-based systems. As a result, the PIM-based system improves performance compared to other computing models due to the short-term data reuse characteristic of computational kernels from PolyBench. However, some kernels perform poorly in PIM-based systems without a multi-layer cache hierarchy due to some kernel's long-term data reuse characteristics. Hence, our evaluation and analysis results suggest that further research should consider dynamic and workload pattern adaptive approaches to overcome performance degradation from computational kernels with long-term data reuse characteristics and hidden data locality.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• Wind and Structures
• /
• 제31권2호
• /
• pp.103-142
• /
• 2020

One of the next frontiers in structural wind engineering is the design of tall buildings using performance-based approaches. Currently, tall buildings are being designed using provisions in the building codes and standards to meet an acceptable level of public safety and serviceability. However, recent studies in wind and earthquake engineering have highlighted the conceptual and practical limitations of the code-oriented design methods. Performance-based wind design (PBWD) is the logical extension of the current wind design approaches to overcome these limitations. Towards the development of PBWD, in this paper, we systematically review the advances made in this field, highlight the research gaps, and provide a basis for future research. Initially, the anatomy of the Wind Loading Chain is presented, in which emphasis was given to the early works of Alan G. Davenport. Next, the current state of practice to design tall buildings for wind load is presented, and its limitations are highlighted. Following this, we critically review the state of development of PBWD. Our review on PBWD covers the existing design frameworks and studies conducted on the nonlinear response of structures under wind loads. Thereafter, to provide a basis for future research, the nonlinear response of simple yielding systems under long-duration turbulent wind loads is studied in two phases. The first phase investigates the issue of damage accumulation in conventional structural systems characterized by elastic-plastic, bilinear, pinching, degrading, and deteriorating hysteretic models. The second phase introduces methods to develop new performance objectives for PBWD based on joint peak and residual deformation demands. In this context, the utility of multi-variate demand modeling using copulas and kernel density estimation techniques is presented. This paper also presents joined fragility curves based on the results of incremental dynamic analysis. Subsequently, the efficiency of tuned mass dampers and self-centering systems in controlling the accumulation of damage in wind-excited structural systems are investigated. The role and the need for explicit modeling of uncertainties in PBWD are also discussed with a case study example. Lastly, two unified PBWD frameworks are proposed by adapting and revisiting the Wind Loading Chain. This paper concludes with a summary and a proposal for future research.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제37권2호
• /
• pp.85-94
• /
• 2010

본 논문에서는 기존의 연구에서 시도되었던 것과는 달리, 복잡하고 추출하기가 어려운 다양한 형태의 자질 및 단서 정보가 필요 없는 합성곱 구문 트리 커널 기반의 단백질 간 상호작용 추출 기법을 소개한다. 이 기법의 특징은 단백질 이름 쌍을 포함한 상호작용 포함 후보 문장에 대한 구문 트리만을 이용하여 추출을 시도한다는 것이며 부가적인 자질이나 커널 함수가 불필요하다는 장점이 있다. 이를 기반으로 본 논문의 연구 성과는 다음과 같다. 첫째, 단백질 간 상호작용 추출에 있어서 구문 트리 커널을 적용할 경우 불필요한 문맥 정보를 효과적으로 제거하는 구문 트리 가지치기 작업이 필수적임을 기존 연구 결과와의 성능 비교로써 증명한다. 둘째, 동일한 학습 조건에서 구문 트리 커널의 소멸 인자(decay factor)는 평활 인자(smoothing factor)로서 중요한 역할을 하며, 성능 변화의 핵심 요소임을 보인다. 특히 학습 집합의 규모에 따라서 소멸인자가 성능에 미치는 영향력이 상이한 패턴으로 나타남을 제시하였다. 결론적으로 기존의 최신 연구결과로서 주장한 "단일 커널보다 혼합 커널의 성능이 더 뛰어나다"라는 가설이 항상 성립하는 것은 아니라는 것을 합성곱 구문 트리 커널 단독으로 적용하여 높은 성능을 나타냄으로써 보여주었다. 동일한 조건으로 수행한 실험에서 기존의 두 연구 결과에 비해 19.8%, 14%의 성능 개선을 나타내었다.