• Interaction and multiscale mechanics
• /
• 제6권2호
• /
• pp.197-210
• /
• 2013

A fundamental trend of processor architecture evolving towards exaflops is fast increasing floating point performance (so-called "free" flops) accompanied by much slowly increasing memory and network bandwidth. In order to fully enjoy the "free" flops, a numerical algorithm of PDEs should request more flops per byte or increase arithmetic intensity. A meshfree/GFEM approximation can be the class of the algorithm. It is shown in a GFEM without extra dof that the kind of approximation takes advantages of the high performance of manycore GPUs by a high accuracy of approximation; the "expensive" method is found to be reversely hardware-efficient on the emerging architecture of manycore.

• 한국음향학회지
• /
• 제3권1호
• /
• pp.25-35
• /
• 1984

This paper analysis the Dynamic Time Warping algorithms for time normalization of speech pattern and discusses the Dynamic Programming algorithm for spoken Korean isolated digits recognition. In the DP matching, feature vectors of the reference and test pattern are consisted of first three formant frequencies extracted by power spectrum density estimation algorithm of the ARMA model. The major differences in the various DTW algorithms include the global path constrains, the local continuity constraints on the path, and the distance weighting/normalization used to give the overall minimum distance. The performance criterias to evaluate these DP algorithms are memory requirement, speed of implementation, and recognition accuracy.

• 재활치료과학
• /
• 제10권1호
• /
• pp.105-114
• /
• 2021

목적 : 본 연구는 시각자극을 사용하여 성별에 따른 일화기억과 공간인지의 차이를 확인하고 비교하기 위해 진행되었다. 연구방법 : 연구에서는 시각적인 자극을 사용한 일화기억 평가와 3차원의 가상현실 기반의 공간인지검사를 사용하여 대학생 48명(남성 24명, 여성 24명)을 대상으로 성별에 따른 일화기억과 공간인지의 차이를 알아보았다. 모든 참가자는 사진을 사용한 시각적 일화기억 평가를 시행하였고 이를 통해 일화기억의 하위 요소인 사물, 장소, 시간 기억에 대한 정확도(%)를 평가하였다. 또한 가상현실에서 길 찾기 과제를 10회 진행하여 가상공간에서의 거리 오차(cm) 값의 평균값을 분석하였다. 결과 : 일화기억은 하위 요소 모두 성별에 따른 유의한 차이가 나타나지 않은 반면(p>.05), 공간인지의 경우 남성이 여성보다 높은 수행을 보이는 것으로 나타났다(p<.05). 결론 : 본 연구의 결과를 통해 성별에 따른 시각적 일화기억의 차이는 언어적 일화기억의 차이와는 다르다는 것을 알 수 있었다. 한편 해마 기능이 일화기억과 공간인지를 담당한다는 것을 고려했을 때, 성별에 따른 일화기억과 공간인지의 차이에 대한 결과가 다르게 나타난 것은 두 기능이 서로 독립적이라는 것을 시사한다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1989년도 봄 학술발표회 논문집
• /
• pp.50-55
• /
• 1989

Piping systems are composed of pipes with various thickness, diameter and length. Accurate analysis of a piping system requires a complicated three dimensional finite element model and a computer system with large memory size, while simplified models result in system response prediction with deteriorated accuracy. An efficient analysis model for piping systems is proposed in this study. The proposed model is developed by introducing a joint model which accounts for the behavior of a pipe connection. Pipes are represented by beam elements and the effect of local deformation of pipe connections are replaced by joint element deformations. The proposed model which is as simple and efficient as a beam model can be used to obtain piping system response with accuracy close to that of a finite element model.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권3호
• /
• pp.193-199
• /
• 2017

Instead of using only spatial features from a single frame for person re-identification, a combination of spatial and temporal factors boosts the performance of the system. A recurrent neural network (RNN) shows its effectiveness in generating highly discriminative sequence-level human representations. In this work, we implement RNN, three Long Short Term Memory (LSTM) network variants, and Gated Recurrent Unit (GRU) on Caffe deep learning framework, and we then conduct experiments to compare performance in terms of size and accuracy for person re-identification. We propose using GRU for the optimized choice as the experimental results show that the GRU achieves the highest accuracy despite having fewer parameters than the others.

• 한국전자파학회논문지
• /
• 제12권2호
• /
• pp.278-284
• /
• 2001

본 논문에서는 기울어진 금속 경계면을 삼각형 격자법으로 해석하고, 계단 근사법과 비교하였다. 특시, 삼각형 격자법에서 금속-유전체간의 경사각도에 대한 오차 범위를 밝혔다. 그결과, 삼각형 격자법에서 경사각이 30。이상 60。이하인 경우, 계단 근사법에 비하여 정확도가 향상되고, 계산속도 및 메로미를 줄일 수 있었다. 그러나 이 범위를 벗어나는 경우, 격자의 한 변의 길이가 상대적으로 늘어나 정확한 해석이 되지 않았다.

• Journal of applied mathematics & informatics
• /
• 제28권1_2호
• /
• pp.439-450
• /
• 2010

In this paper, we propose a new idea to evaluate the prediction accuracy of user's preference generated by memory-based collaborative filtering algorithm before prediction process in the recommender system. Our analysis results show the possibility of a pre-evaluation before the prediction process of users' preference of item's transaction on the web. Classification functions proposed in this study generate a user's rating pattern under certain conditions. In this research, we test whether classification functions select users who have lower prediction or higher prediction performance under collaborative filtering recommendation approach. The statistical test results will be based on the differences of the prediction accuracy of each user group which are classified by classification functions using the generative probability of specific rating. The characteristics of rating patterns of classified users will also be presented.

• 소음진동
• /
• 제2권4호
• /
• pp.265-272
• /
• 1992

The simplest method which has been used extensively for vibration analysis is the transfer matrix method introduced by Myklestad and was later extended by many researchers. The crude approximation results in considerable error on the predicted natural frequencies and to increase the accuracy the number of elements used in the analysis must be increased. In addition, numerical instability can occur as a result of matrix multiplication. Also the main disadvantage of the finite element method is the large computer memory requirements for complex systems. The new method proposed in this paper combines the transfer matrix and finite dynamic element techniques to form a powerful algorithm for vibration analysis of rotor system. It is shown that the accuracy improves significantly when the transfer matrix for each segment is obtained from finite dynamic element techniques.

• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• 한국컴퓨터정보학회논문지
• /
• 제23권11호
• /
• pp.31-41
• /
• 2018

Text classification has been studied for a long time in the Natural Language Processing field. In this paper, we propose an article- and paragraph-level genre classification system using Word2Vec-based LSTM, GRU, and CNN models for large-scale English corpora. Both article- and paragraph-level classification performed best in accuracy with LSTM, which was followed by GRU and CNN in accuracy performance. Thus, it is to be confirmed that in evaluating the classification performance of LSTM, GRU, and CNN, the word sequential information for articles is better than the word feature extraction for paragraphs when the pre-trained Word2Vec-based word embeddings are used in both deep learning-based article- and paragraph-level classification tasks.