• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.83-86
• /
• 2017

단일 코어 프로세스의 성능 향상은 전력 소모, 발열 등의 이유로 한계에 달했다. 이에 대한 대안으로 멀티 코어가 등장했으며 매니 코어 기술에 대한 연구가 활발히 진행 중에 있다. 이렇듯 멀티 코어 환경이 보편화됨에 따라 병렬 프로그래밍의 중요성이 더욱 커졌다. 한편, 순수 함수형 언어 Haskell은 부수효과가 없고 다양한 병렬화 도구를 지원함으로써 다가오는 병렬 프로그래밍 시대에 적합한 언어라 할 수 있다. 이때 Haskell 병렬 프로그램의 성능은 메모리 재사용(Garbage Collection) 시간에 큰 영향을 받는다. 그래서 Haskell 병렬 프로그램의 성능 향상, 분석을 위한 메모리 프로파일링 도구가 필요하다. 이미 Haskell이 제공하는 메모리 프로파일링 도구로 ghc-gc-tune이 있지만 실행 속도 측면에서 개선이 필요하다. 본 연구에서는 분할 정복법을 이용해서 매 단계마다 탐색 영역을 4분의 1로 줄이도록 ghc-gc-tune을 개선했다. 개선된 ghc-gc-tune을 극대 독립 집합 프로그램과 K-means 프로그램에 적용한 결과, 평균 98%의 정확도로 실행 시간을 평균 7.78배 단축했다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.427-431
• /
• 2010

Multidimensional scaling (MDS) is a widely used method for dimensionality reduction, of which purpose is to represent high-dimensional data in a low-dimensional space while preserving distances among objects as much as possible. MDS has mainly been applied to data visualization and feature selection. Among various MDS methods, the classical MDS is not readily applicable to data which has large numbers of objects, on normal desktop computers due to its computational complexity. More precisely, it needs to solve eigenpair problems on dissimilarity matrices based on Euclidean distance. Thus, running time and required memory of the classical MDS highly increase as n (the number of objects) grows up, restricting its use in large-scale domains. In this paper, we propose an efficient approximation algorithm for the classical MDS based on divide-and-conquer and CUDA. Through a set of experiments, we show that our approach is highly efficient and effective for analysis and visualization of data consisting of several thousands of objects.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.9
• /
• pp.377-384
• /
• 2017

The performance improvement of a single core processor has reached its limit since the circuit density cannot be increased any longer due to overheating. Therefore, the multicore and manycore architectures have emerged as viable approaches and parallel programming becomes more important. Haskell, a purely functional language, is getting popular in this situation since it naturally supports parallel programming owing to its beneficial features including the implicit parallelism in evaluating expressions and the monadic tools supporting parallel constructs. However, the performance of Haskell parallel programs is strongly influenced by the performance of the run-time system including the garbage collector. Though a memory profiling tool namely GC-tune has been suggested, we need a more systematic way to use this tool. Since GC-tune finds the optimal memory size by executing the target program with all the different possible GC options, the GC-tuning time takes too long. This paper suggests a basic divide-and-conquer method to reduce the number of GC-tune executions by reducing the search area by one-quarter for every searching step. Applying this method to two parallel programs, a maximally independent set and a K-means programs, the memory tuning time is reduced by 7.78 times with accuracy 98% on average.

• The Korean Journal of Applied Statistics
• /
• v.33 no.5
• /
• pp.569-578
• /
• 2020

By estimating conditional quantile functions of the response, quantile regression (QR) can provide comprehensive information of the relationship between the response and the predictors. In addition, kernel quantile regression (KQR) estimates a nonlinear conditional quantile function in reproducing kernel Hilbert spaces generated by a positive definite kernel function. However, it is infeasible to use the KQR in analysing a massive data due to the limitations of computer primary memory. We propose a divide and conquer based KQR (DC-KQR) method to overcome such a limitation. The proposed DC-KQR divides the entire data into a few subsets, then applies the KQR onto each subsets and derives a final estimator by aggregating all results from subsets. Simulation studies are presented to demonstrate the satisfactory performance of the proposed method.

• The Korean Journal of Applied Statistics
• /
• v.35 no.2
• /
• pp.177-188
• /
• 2022

The support vector machine (SVM) has been successfully applied to various classification areas with a high level of classification accuracy. However, it is infeasible to use the SVM in analyzing massive data because of its significant computational problems. When analyzing imbalanced data with different class sizes, furthermore, the classification accuracy of SVM in minority class may drop significantly because its classifier could be biased toward the majority class. To overcome such a problem, we propose the DOC-SVM method, which uses divide-oversampling and conquers techniques. The proposed DOC-SVM divides the majority class into a few subsets and applies an oversampling technique to the minority class in order to produce the balanced subsets. And then the DOC-SVM obtains the final classifier by aggregating all SVM classifiers obtained from the balanced subsets. Simulation studies are presented to demonstrate the satisfactory performance of the proposed method.

• Clinics in Shoulder and Elbow
• /
• v.16 no.1
• /
• pp.17-26
• /
• 2013

Purpose: A comparison of the radiographic and the clinical outcomes between two different surgical approaches-Deltoid splitting and Delto-pectoral interval-on the proximal humerus fractures treated by locking compressive plate (LCP), is done. Materials and Methods: Medical records and pre- and postoperative radiographs were reviewed retrospectively for 75 adult patients who underwent surgical fixations with locking compressive plates from May 2005 to December 2011. Patients were divided into two groups according to the surgical methods. Differences in the neck-shaft angle between immediate postoperative period and final follow-up were compared between the two groups. Differences in constant score and Korean shoulder score (KSS) between affected arms and contralateral arms at final follow-up were also compared. Results: The differences in the neck-shaft angle between immediate postoperative period and at final follow-up was 12.04 degrees on average in Deltoid splitting approach and 10.20 degrees in Delto-pectoral interval approach, which was not statistically significant. Differences in constant score/KSS between the affected arm and the contralateral arm were 13.78/22.74 points in deltoid-splitting approach on average and 19.41/31.13 points in Delto-pectoral interval approach, showing that deltoid-splitting approach is significantly superior. Conclusion: Deltoid-splitting approach showed better functional outcomes in the fracture reduction and internal fixation using LCP for the treatment of unstable proximal humerus fractures.

• The KIPS Transactions:PartA
• /
• v.8A no.1
• /
• pp.36-41
• /
• 2001

In this paper, we first design an parallel quadratic sieve algorithm for factoring method. We then present parallel factoring algorithm for factoring a large odd integer by repeatedly using the parallel quadratic sieve algorithm based on the divide-and-conquer strategy on SIMD machines with DMM. We show that this algorithm is optimal in view of the product of time and processor numbers.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.1-10
• /
• 2021

The ability to recognize human emotions by computer vision is a very important task, with many potential applications. Therefore the demand for emotion recognition using not only RGB images but also thermal images is increasing. Compared to RGB images, thermal images has the advantage of being less affected by lighting conditions but require a more sophisticated recognition method with low-resolution sources. In this paper, we propose a Divide and Conquer-based CNN training strategy to improve the performance of facial thermal image-based emotion recognition. The proposed method first trains to classify difficult-to-classify similar emotion classes into the same class group by confusion matrix analysis and then divides and solves the problem so that the emotion group classified into the same class group is recognized again as actual emotions. In experiments, the proposed method has improved accuracy in all the tests than when recognizing all the presented emotions with a single CNN model.

• Journal of KIISE:Software and Applications
• /
• v.26 no.2
• /
• pp.273-283
• /
• 1999

래스터 지도에서 직선 또는 곡선과 중첩되어 있는 경우의 문자는 추출하기가 쉽지 않다. 따라서 본 논문에서는 고립되어 있는 문자뿐만 아니라 문자이외의 요소와 중첩되어 있는 문자도 효과적으로 추출할수 있는 분할 정복(divide and conquer) 개념에 기반한 문자 추출방법을 제시한다. 이를 위해 먼저 이미지의 연결 요소로부터 볼록다각형(convex hull)을 생성한다. 그리고 이 다각형이 충분한게 문자영역만을 포함할때가지 볼록 다각형을 이등분하면서 가장 긴 선분(투사 선분)을 기준으로 두 영역으로 분할한다. 다음으로 문자를 추출하기 위해서 이 선분을 기준으로 연결 요소상의 픽셀의 밀집도를 계산하는 알고리즘(프로파일링)을 적용한다. 또한 지도상에서 추출된 개별적인 문자들을 의미있는 단어들로 묶기(grouping)한 새로운 알고리즘을 소개한다. 특히 지도상에 나타나는 문자의 종류는 매우 다양하고 또한 이 문자들이 놓여있는 방향 역시 일정하지 않기 때문에 이러한 단어를 찾는 kd법은 쉽지 않다. 이를 위해 본 논문에서는 3차원 인접 그래프(3-D neighborhood graph)G를 소개한다. 이 그래프 G에서 각 노드는 하나의 분리된 문자를 나타내며 자신의 크기와 위치에 따라서 3차원 공간상에서 위치하게된다. 따라서, 크기가 큰 (작은)문자들은 보다 큰 (작은) z값을 가지고 되며 이 그래프 G에서 서로 인접한 노드들을 연결함으로써 지도상에 존재하는 서로 다른 종류의 문자 스트링을 추출할수 있다. 실험결과는 서로 다른 지도 이미지에 대해서 약 95% 이상의 단어 추출율을 보여준다.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.4
• /
• pp.69-78
• /
• 2023

With the rapid growth of artificial intelligence, the demand for semiconductors is enormously increasing everywhere. To ensure the manufacturing quality and quantity simultaneously, the importance of automatic defect detection during the packaging process has been re-visited by adapting various deep learning-based methodologies into automatic packaging defect inspection. Deep learning (DL) models require a large amount of data for training, but due to the nature of the semiconductor industry where security is important, sharing and labeling of relevant data is challenging, making it difficult for model training. In this study, we propose a new framework for securing sufficient data for DL models with fewer computing resources through a divide-and-conquer approach. The proposed method divides high-resolution images into pre-defined sub-regions and assigns conditional labels to each region, then trains individual sub-regions and boundaries with boundary loss inducing the globally coherent and seamless images. Afterwards, full-size image is reconstructed by combining divided sub-regions. The experimental results show that the images obtained through this research have high efficiency, consistency, quality, and generality.