• 대한수학회논문집
• /
• 제19권1호
• /
• pp.179-196
• /
• 2004

Mond-Weir type duals for multiobjective variational problems are formulated. Under generalized vector variational type I invexity assumptions on the functions involved, sufficient optimality conditions, weak and strong duality theorems are proved efficient and properly efficient solutions of the primal and dual problems.

• 한국음향학회지
• /
• 제39권2호
• /
• pp.137-142
• /
• 2020

문장 종속 짧은 발화에서 문장 독립 긴 발화까지 다양한 환경에서 I-vector 특징에 기반을 둔 많은 연구가 수행되었다. 본 논문에서는 원거리 잡음 환경에서 녹음한 데이터에서 Probabilistic Linear Discriminant Analysis(PLDA)를 적용한 I-vector와 주의 집중 기법을 접목한 Long Short Term Memory(LSTM) 기반의 화자 임베딩을 추출하여 결합한 화자 검증 알고리즘을 소개한다. LSTM 모델의 Equal Error Rate(EER)이 15.52 %, Attention-LSTM 모델이 8.46 %로 7.06 % 성능이 향상되었다. 이로써 본 논문에서 제안한 기법이 임베딩을 휴리스틱 하게 정의하여 사용하는 기존 추출방법의 문제점을 해결할 수 있는 것을 확인하였다. PLDA를 적용한 I-vector의 EER이 6.18 %로 결합 전 가장 좋은 성능을 보였다. Attention-LSTM 기반 임베딩과 결합하였을 때 EER이 2.57 %로 기존보다 3.61 % 감소하여 상대적으로 58.41 % 성능이 향상되었다.

• 한국통신학회논문지
• /
• 제41권8호
• /
• pp.958-964
• /
• 2016

본 논문에서는 여러 개의 이원 support vector machine (binary SVM)을 사용하여 세 개 이상의 클래스를 분류하는 multi-class SVM과 유사하게 다중의 판별 deep neural network (DNN) 모델을 사용하는 i-벡터 기반의 언어 인식 시스템을 제안한다. 제안하는 시스템은 NIST 2015 i-vector Machine Learning Challenge 데이터베이스에 포함된 i-벡터들을 이용하여 학습 및 테스트 되었으며, 오픈 세트에서 기존의 cosine distance, multi-class SVM 및 단일 neural network (NN) 기반의 언어 인식 시스템에 비하여 높은 성능을 보임이 확인되었다.

• 대한수학회보
• /
• 제37권2호
• /
• pp.229-247
• /
• 2000

This paper is concerned with the impulsive Cauchy problem where the control function u is a possibly discontinuous vector-valued function with finite total variation. We assume that the vector fields f, $g_i$(i=1,…, m) are dependent on the time variable. The impulsive Cauchy problem is of the form x(t)=f(t,x) +$\SUMg_i(t,x)u_i(t)$, $t\in$[0,T], x(0)=$\in\; R^n$, where the vector fields f, $g_i$ : $\mathbb{R}\; \times\; \mathbb{R}\; \longrightarrow\; \mathbb(R)^n$ are measurable in t and Lipschitz continuous in x, If $g_i's$ satisfy a condition that $\SUM{\mid}g_i(t_2,x){\mid}{\leq}{\phi}$ $\forallt_1\; <\; t-2,x\; {\epsilon}\;\mathbb{R}^n$ for some increasing function $\phi$, then the imput-output function can be continuously extended to measurable functions of bounded variation.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 1998년도 연구회 합동 학술발표회 논문집
• /
• pp.5-8
• /
• 1998

The adaptation to vector control theory is so generalized that it is widely used for implementing the high-performance of AC machine. Nowadays, One-Chip microprocessors or DSP chips are being well-used to implement Vector Control algorithm. DSP Chip have less flexibility for memory decoding and I/O rather than One-Chip microprocessor so that is requires more additional circuit and high cost. And the past One-Chip micro processors have difficult of implementation the complex algorithm because of small memory capacity and low arithmetic performance. Therefore we implemented the vector control algorithm of PMSM(Permanent Magnetic Synchronous Motors) using 80296SA form intel , which have many features as 6M memory space, 500MHz clock frequency, including memory decoding circuit and general I/O, Special I/O(EPA, Interrupt controller, Timer/Count, PWM generator) which is proper controller for the complex algorithm or operation program requiring so much memory capacity, So in this paper we fully digitized the vector control of PMSM included SVPWM Voltage controller using the intel 80296SA

• 대한수학회논문집
• /
• 제23권2호
• /
• pp.191-199
• /
• 2008

We investigate the relation between the vector variable bi-additive functional equation $f(\sum\limits^n_{i=1} xi,\;\sum\limits^n_{i=1} yj)={\sum\limits^n_{i=1}\sum\limits^n_ {j=1}f(x_i,y_j)$ and the multi-variable quadratic functional equation $$g(\sum\limits^n_{i=1}xi)\;+\;\sum\limits_{1{\leq}i<j{\leq}n}\;g(x_i-x_j)=n\sum\limits^n_{i=1}\;g(x_i)$$. Furthermore, we find out the general solution of the above two functional equations.

• Computers and Concrete
• /
• 제5권5호
• /
• pp.461-473
• /
• 2008

The paper explores the potential of Support Vector Machines (SVM) approach in predicting 28-day compressive strength and slump flow of self-compacting concrete. Total of 80 data collected from the exiting literature were used in present work. To compare the performance of the technique, prediction was also done using a back propagation neural network model. For this data-set, RBF kernel worked well in comparison to polynomial kernel based support vector machines and provide a root mean square error of 4.688 (MPa) (correlation coefficient=0.942) for 28-day compressive strength prediction and a root mean square error of 7.825 cm (correlation coefficient=0.931) for slump flow. Results obtained for RMSE and correlation coefficient suggested a comparable performance by Support Vector Machine approach to neural network approach for both 28-day compressive strength and slump flow prediction.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2011년도 제42회 하계학술대회
• /
• pp.2075-2076
• /
• 2011

The stator winding faults diagnosis technique based on MCSA is as follows. Firstly, collecting the 3 phase motor currents, that signal is transformed by (d-q transform, $i_d$, $i_q$). Park's vector pattern, the circle that is down by d-q transformed currents($i_d$, $i_q$). The circle is widely used for stator winding faults detection. The current distortion ratio(DR), defined by the ratio of max-axis and min-axis of ellipse of Park's vector's pattern. In this study, distortion ratio of Park's vector pattern is suggested for Auto diagnosis of stator winding short fault and usefulness of distortion ratio is verified through simulation using LabVIEW program.

• Journal of Microbiology
• /
• 제44권6호
• /
• pp.671-673
• /
• 2006

A new 4.87 kb Escherichia-Pseudomonas shuttle vector has been constructed by inserting a 1.27 kb DNA fragment with a replication origin of a Pseudomonas plasmid pRO1614 into the 3.6 kb E. coli plasmid pBGS18. This vector, designated pJH1, contains an aminogly-coside phosphotransferase gene (aph) from Tn903, a lacZ' gene for $\alpha$-complementation and a versatile multiple cloning site possessing unique restriction sites for EcoRI, SacI, KpnI, SmaI, BamHI, XbaI, SalI, BspMI, PstI, SphI, and HindIII. When pJH1 was transformed into E. coli DHS${\alpha}$ and into P. putida HK-6, it was episomally and stably maintained in both strains. In addition, the enhanced green fluorescent protein (EGFP) gene which was transcriptionally cloned into pJH1 rendered E. coli cells fluorescence when its transformants were illuminated at 488 nm.

• Biomolecules & Therapeutics
• /
• 제10권1호
• /
• pp.37-42
• /
• 2002

Brain drug targeting through the blood-brain barrier (BBB) in vivo is possible with peptidornirnetic monoclonal antibodies that undergo receptor-mediated transcytosis through the BBB. Monoclonal antibody to the rat transferrin receptor, such as the OX26 was studied in rats as a transport vector through BBB on the transferrin receptor. But, OX26 is not an effective brain delivery vector in mouse. In the present studies, rat monoclonal antibody, 8D3 to the mouse transferrin receptor were evaluated for brain drug targeting vector intransgenic mouse model. Pharrnacokinetic parameters in plasma and organ uptakes were determined at varioustimes after i.v. bolus injection of [$^{}125}I$] 8D3 in Balb/c mice. Brain uptake of [$^{}125}I$] 8D3 was also studied with an internal carotid artery perfusioncapillary depletion method. After i.v. injection of [$^{}125}I$] 8D3, plasma concentrations declined biexponentially with elimination half lift of approximately 2.2 hours. Brain uptake of [$^{}125}I$] 8D3 was $0.50{\pm}0.09$ persent of injected dose per g brain after 2 hours i.v. injection. After perfusion 5 min the apparent volume of distibution of [$^{}125}I$] 8D3 in brain was $22.3 {\mu}l/g,$ which was 4.8 fold higher than the intravascular volume. These studies indicate rat monoclonal antibody to the mouse transferrin receptor, 8D3 may be used for brain drug targeting vector in mice.