• Journal of Multimedia Information System
• /
• 제2권2호
• /
• pp.215-222
• /
• 2015

Epileptic Seizure is a popular brain disease in the world. It affects the nervous system and the activities of brain function that make a person who has seizure signs cannot control and predict his actions. Based on the Electroencephalography (EEG) signals which are recorded from human or animal brains, the scientists use many methods to detect and recognize the abnormal activities of brain. Tucker model is investigated to solve this problem. Tucker decomposition is known as a higher-order form of Singular Value Decomposition (SVD), a well-known algorithm for decomposing a matric. It is widely used to extract good features of a tensor. After decomposing, the result of Tucker decomposition is a core tensor and some factor matrices along each mode. This core tensor contains a number of the best information of original data. In this paper, we used Tucker decomposition as a way to obtain good features. Training data is primarily applied into the core tensor and the remained matrices will be combined with the test data to build the Tucker base that is used for testing. Using core tensor makes the process simpler and obtains higher accuracies.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제14권3호
• /
• pp.296-300
• /
• 2008

최근에 개발된 Nonnegative tensor factorization(NTF)는 비음수 행렬 분해(NMF)의 multiway(multilinear) 확장형이다. NTF는 CANDECOMP/PARAFAC 모델에 비음수 제약을 가한 모델이다. 본 논문에서는 Tucker 모델에 비음수 제약을 가한 nonnegative Tucker decomposition(NTD)라는 새로운 텐서 분해 모델을 제안한다. 제안된 NTD 모델을 least squares, I-divergence, $\alpha$-divergence를 이용한 여러 목적함수에 대하여 fitting하는 multiplicative update rule을 유도하였다.

• 대한수학회지
• /
• 제54권3호
• /
• pp.807-820
• /
• 2017

An open smooth manifold is said of finite geometric rank if it admits a handlebody decomposition with a finite number of 1-handles. We prove that, if there exists a proper submanifold $W^{n+3}$ of finite geometric rank between an open 3-manifold $V^3$ and its stabilization $V^3{\times}B^n$(where $B^n$ denotes the standard n-ball), then the manifold $V^3$ has the Tucker property. This means that for any compact submanifold $C{\subset}V^3$, the fundamental group ${\pi}_1(V^3-C)$ is finitely generated. In the irreducible case this implies that $V^3$ has a well-behaved compactification.

• International Journal of Contents
• /
• 제12권3호
• /
• pp.22-28
• /
• 2016

Tensor with missing or incomplete values is a ubiquitous problem in various fields such as biomedical signal processing, image processing, and social network analysis. In this paper, we considered how to reconstruct a dataset with missing values by using tensor form which is called tensor completion process. We applied Tucker factorization to solve tensor completion which was built base on optimization problem. We formulated the optimization objective function using components of Tucker model after decomposing. The weighted least square matric contained only known values of the tensor with low rank in its modes. A first order optimization method, namely Nonlinear Conjugated Gradient, was applied to solve the optimization problem. We demonstrated the effectiveness of the proposed method in EEG signals with about 70% missing entries compared to other algorithms. The relative error was proposed to compare the difference between original tensor and the process output.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2007년도 가을 학술발표논문집 Vol.34 No.2 (A)
• /
• pp.209-210
• /
• 2007

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.657-667
• /
• 2019

This paper studies synthetic aperture radar (SAR) imaging problem which the scatterers are often distributed in block sparse pattern. To exploiting the sparse geometrical feature, a Kronecker constrained SAR imaging algorithm is proposed by combining the block sparse characteristics with the multiway sparse reconstruction framework with Tucker modeling. We validate the proposed algorithm via real data and it shows that the our algorithm can achieve better accuracy and convergence than the reference methods even in the demanding environment. Meanwhile, the complexity is smaller than that of the existing methods. The simulation experiments confirmed the effectiveness of the algorithm as well.

• 한국정보통신학회논문지
• /
• 제26권6호
• /
• pp.842-849
• /
• 2022

딥러닝 기술의 발달로 무인 자동차, 드론, 로봇 등의 임베디드 시스템 분야에서 DNN을 활용하는 사례가 많아지고 있다. 대표적으로 자율주행 시스템의 경우 정확도가 높고 연산량이 큰 몇 개의 DNN들을 동시에 수행하는 것이 필수적이다. 하지만 상대적으로 낮은 성능을 갖는 임베디드 환경에서 다수의 DNN을 동시에 수행하면 추론에 걸리는 시간이 길어진다. 이러한 현상은 추론 결과에 따른 동작이 제때 이루어지지 않아 비정상적인 기능을 수행하는 문제를 발생시킬 수 있다. 이를 해결하기 위하여 본 논문에서 제안한 솔루션은 먼저 연산량이 큰 DNN에 터커 분해 기법을 적용하여 연산량을 감소시킨다. 그다음으로 DNN 모델들을 GPU 내부에서 은닉층 단위로 최대한 병렬적으로 수행될 수 있게 한다. 실험 결과 DNN의 추론 시간이 제안된 기법을 적용하기 전 대비 최대 75.6% 감소하였다.

• Journal of Communications and Networks
• /
• 제13권6호
• /
• pp.633-638
• /
• 2011

In this paper, we consider resource allocation with proportional fairness in the downlink orthogonal frequency division multiple access relay networks, in which relay nodes operate in decode-and-forward mode. A joint optimization problem is formulated for relay selection, subcarrier assignment and power allocation. Since the formulated primal problem is nondeterministic polynomial time-complete, we make continuous relaxation and solve the dual problem by Lagrangian dual decomposition method. A near-optimal solution is obtained using Karush-Kuhn-Tucker conditions. Simulation results show that the proposed algorithm provides superior system throughput and much better fairness among users comparing with a heuristic algorithm.