• 정보처리학회논문지D
• /
• 제19D권2호
• /
• pp.147-150
• /
• 2012

클래스들 간의 거리를 최대화시키는 사영 방향을 구하는 감독차원감소 방법인 선형판별분석법(LDA)은 클래스 정보를 가진 데이터의 수가 매우 적을 때 성능이 급격히 저하되는 경향이 있다. 이러한 경우 상대적으로 저렴한 비용으로 얻을 수 있는 클래스 라벨 정보가 없는 데이터를 활용할 수 있는 반감독 차원 감소법이 사용될 수 있다. 그러나 통계적 차원 감소법에서 흔히 사용되는 행렬연산은 많은 양의 데이터를 사용하는데 메모리와 처리시간에서 한계가 있고, 적은 수의 라벨드 데이터(labeled data)에 비해 너무나 많은 언라벨드 데이터(unlabeled data)의 사용은 처리 시간의 증가에 비해 오히려 성능감소를 가져올 수 있다. 이러한 문제들을 극복하기 위해 앙상블 접근법을 이용한 반감독 차원 감소 방법을 제안한다. 문서분류 문제에서의 실험결과를 통해 제안한 방법의 성능을 입증한다.

• 응용통계연구
• /
• 제36권4호
• /
• pp.323-335
• /
• 2023

고차원 데이터의 인과 추론에서 고차원 공변량의 차원을 축소하고 적절히 변형하여 처리와 잠재 결과에 영향을 줄 수 있는 교란을 통제하는 것은 중요한 문제이다. 평균 처리 효과(average treatment effect; ATE) 추정에 있어서, 성향점수와 결과 모형 추정을 이용한 확장된 역확률 가중치 방법이 주로 사용된다. 고차원 데이터의 분석시 모든 공변량을 포함한 모수 모형을 이용하여 성향 점수와 결과 모형 추정을 할 경우, ATE 추정량이 일치성을 갖지 않거나 추정량의 분산이 큰 값을 가질 수 있다. 이런 이유로 고차원 데이터에 대한 적절한 차원 축소 방법과 준모수 모형을 이용한 ATE 방법이 주목 받고 있다. 이와 관련된 연구로는 차원 축소부분에 준모수 모형과 희소 충분 차원 축소 방법을 활용한 연구가 있다. 최근에는 성향점수와 결과 모형을 추정하지 않고, 차원 축소 후 매칭을 활용한 ATE 추정 방법도 제시되었다. 고차원 데이터의 ATE 추정 방법연구 중 최근에 제시된 네 가지 연구에 대해 소개하고, 추정치 해석시 유의할 점에 대하여 논하기로 한다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2007년도 정기 학술대회 논문집
• /
• pp.135-140
• /
• 2007

In this paper, an Improved Dimension Reduction(IDR) method is proposed for uncertainty quantification that employes Kriging interpolation technic. It has been acknowledged that the DR method is accurate and efficient for assessing statistical moments and reliability due to the sensitivity free feature. However, the DR method has a number of drawbacks such as instability and inaccuracy for problems with increased nonlineality. In this paper, improved DR is implanted by three steps. First, the Kriging interpolation method is used to accurately approximate the responses. Second, 2N+1 and 4N+1 ADOEs are proposed to maintain high accuracy of the method for UQ analysis. Third, numerical integration scheme is used with accurate but free response values at any set of integration points of the surrogated model.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.107-115
• /
• 2012

본 논문에서는 Yoo와 Cook (2007)에 의하여 제시된 다변량 회귀의 조건부 평균에 대한 최소 불일치 함수 접근법을 통한 최적 차원축소 부분공간의 추정에서 차원의 수가 추정된 선형결합들과 설명력 등에 어떤 영향을 미치는 지를 시뮬레이션 자료를 통하여 알아보았다. 그 결과 추정에 사용된 차원수에 따른 여러 결과들을 차원결정을 위한 검정과 함께 활용하면 모형에 필요한 차원수를 탐색하는데 매우 효과적임을 알 수 있었다.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• pp.567-576
• /
• 2006

In this paper, we applied the multiblock dimension reduction methods to the classification of tumor based on microarray gene expressions data. This procedure involves clustering selected genes, multiblock dimension reduction and classification using linear discrimination analysis and quadratic discrimination analysis.

• 말소리와 음성과학
• /
• 제7권4호
• /
• pp.3-8
• /
• 2015

In this paper, we investigate an input dimension reduction method using continuous word vector in deep neural network language model. In the proposed method, continuous word vectors were generated by using Google's Word2Vec from a large training corpus to satisfy distributional hypothesis. 1-of-${\left|V\right|}$ coding discrete word vectors were replaced with their corresponding continuous word vectors. In our implementation, the input dimension was successfully reduced from 20,000 to 600 when a tri-gram language model is used with a vocabulary of 20,000 words. The total amount of time in training was reduced from 30 days to 14 days for Wall Street Journal training corpus (corpus length: 37M words).

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2008년도 정기 학술대회
• /
• pp.422-427
• /
• 2008

This study is to illustrate the usefulness of Kriging Dimension Reduction Method(KDRM), which is to construct probability distribution of response function in the presence of the physical uncertainty of input variables. DRM has recently received increased attention due to its sensitivity-free nature and efficiency that considerable accuracy is obtained with only a few number of analyses. However, the DRM has a number of drawbacks such as instability and inaccuracy for functions with increased nonlinearity. As a remedy, Kriging interpolation technique is incorporated which is known as more accurate for nonlinear functions. The KDRM is applied and compared with MCS methods in a compression coil spring design problem. The effectiveness and accuracy of this method is verified.

• 대한기계학회논문집A
• /
• 제33권9호
• /
• pp.870-877
• /
• 2009

This study provides how the Dimension Reduction (DR) method as an efficient technique for reliability analysis can acquire its increased efficiency when it is applied to highly nonlinear problems. In the highly nonlinear engineering systems, 4N+1 (N: number of random variables) sampling is generally recognized to be appropriate. However, there exists uncertainty concerning the standard for judgment of non-linearity of the system as well as possibility of diverse degrees of non-linearity according to each of the random variables. In this regard, this study judged the linearity individually on each random variable after 2N+1 sampling. If high non-linearity appeared, 2 additional sampling was administered on each random variable to apply the DR method. The applications of the proposed sampling to the examples produced the constant results with increased efficiency.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.832-854
• /
• 2019

Action recognition has been studied in computer vision field for years. We present an effective approach to recognize actions using a dimension reduction method, which is applied as a crucial step to reduce the dimensionality of feature descriptors after extracting features. We propose to use sparse matrix and randomized kd-tree to modify it and then propose modified Local Fisher Discriminant Analysis (mLFDA) method which greatly reduces the required memory and accelerate the standard Local Fisher Discriminant Analysis. For feature encoding, we propose a useful encoding method called mix encoding which combines Fisher vector encoding and locality-constrained linear coding to get the final video representations. In order to add more meaningful features to the process of action recognition, the convolutional neural network is utilized and combined with mix encoding to produce the deep network feature. Experimental results show that our algorithm is a competitive method on KTH dataset, HMDB51 dataset and UCF101 dataset when combining all these methods.

• Communications for Statistical Applications and Methods
• /
• 제24권3호
• /
• pp.303-315
• /
• 2017

Yoo (2015, Statistics and Probability Letters, 99, 109-113) derives theoretical results in an optimal sufficient dimension reduction with singular inner-product matrix. The results are promising, but Yoo (2015) only presents one simulation study. So, an evaluation of its practical usefulness is necessary based on numerical studies. This paper studies the asymptotic behaviors of Yoo (2015) through various simulation models and presents a real data example that focuses on ordinary least squares. Intensive numerical studies show that the $x^2$ test by Yoo (2015) outperforms the existing optimal sufficient dimension reduction method. The basis estimation by the former can be theoretically sub-optimal; however, there are no notable differences from that by the latter. This investigation confirms the practical usefulness of Yoo (2015).