• Title/Summary/Keyword: multiple weights

Search Result 298, Processing Time 0.022 seconds

Learning-Based Multiple Pooling Fusion in Multi-View Convolutional Neural Network for 3D Model Classification and Retrieval

  • Zeng, Hui;Wang, Qi;Li, Chen;Song, Wei
    • Journal of Information Processing Systems
    • /
    • v.15 no.5
    • /
    • pp.1179-1191
    • /
    • 2019
  • We design an ingenious view-pooling method named learning-based multiple pooling fusion (LMPF), and apply it to multi-view convolutional neural network (MVCNN) for 3D model classification or retrieval. By this means, multi-view feature maps projected from a 3D model can be compiled as a simple and effective feature descriptor. The LMPF method fuses the max pooling method and the mean pooling method by learning a set of optimal weights. Compared with the hand-crafted approaches such as max pooling and mean pooling, the LMPF method can decrease the information loss effectively because of its "learning" ability. Experiments on ModelNet40 dataset and McGill dataset are presented and the results verify that LMPF can outperform those previous methods to a great extent.

An Efficient Diagnosis Algorithm for Multiple Stuck-at Faults (다중 고착 고장을 위한 효율적인 고장 진단 알고리듬)

  • Lim Yo-Seop;Lee Joo-Hwan;Kang Sung-Ho
    • Journal of the Institute of Electronics Engineers of Korea SD
    • /
    • v.43 no.9 s.351
    • /
    • pp.59-63
    • /
    • 2006
  • With the increasing complexity of VLSI devices, more complex faults have appeared. Many methods for diagnosing the single stuck-at fault have been studied. Often multiple defects on a foiling chip better reflect the reality. So, we propose an efficient diagnosis algorithm for multiple stuck-at faults. By using vectorwise intersections as an important metric of diagnosis, the proposed algorithm can diagnose multiple defects using single stuck-at fault simulator. In spite of multiple fault diagnosis, the number of candidate faults is also drastically reduced. For fault identification, positive calculations and negative calculations based on variable weights are used for the matching algorithm. Experimental results for ISCAS85 and full-scan version of ISCAS89 benchmark circuits prove the efficiency of the proposed algorithm.

DWTHE: Decomposable Weighted and Thresholded Histogram Equalization (DWTHE: 분할 기반의 히스토그램 평활화)

  • Kim, Mary;Chung, Min-Gyo
    • Journal of KIISE:Computing Practices and Letters
    • /
    • v.15 no.11
    • /
    • pp.856-860
    • /
    • 2009
  • Wang and Ward developed an image contrast enhancement method called WTHE (Weighted and Thresholded Histogram Equalization). In this paper, we propose an improved variant of WTHE called DWTHE(Decomposable WTHE) that enhances WTHE through the use of histogram decomposition. Specifically, DWTHE divides an input histogram by using image's mean brightness or equally-spaced brightness points, computes a probability value for each sub-histogram, modifies the sub-histograms by using those probability values as weights, and then performs histogram equalization on the modified input histogram. As opposed to WTHE that uses a single weight, DWTHE uses multiple weights derived from histogram decomposition. Experimental results show that the proposed method outperforms the previous histogram equalization based methods.

An Analysis of the Economic Sensitivity of Imported Fishery Products (수입수산물의 경제적 민감도분석에 관한 연구)

  • Park, Cheol-Hyung;Jang, Young-Soo
    • Journal of Fisheries and Marine Sciences Education
    • /
    • v.20 no.1
    • /
    • pp.78-89
    • /
    • 2008
  • This study is intended to analyse the economic sensitivity of imported fishery products due to decrease in or elimination of tariff rates through the progress of free trade. Forty-seven species of fishes were selected for this study on the basis of the HS Code. The substitution and price effects were calculated using the price elasticities of both domestic and imported demands for fishery products under the assumption of 5% decrease in a tariff rate. Seven main economic variables were extracted from the fishery industry which can mediate the substitution and price effects. A multiple regression analysis was conducted to obtain the influence weights of these main economic variables on both effects. The order of sensitivity of the fishes was calculated using these weights. The 47 fish species were classified into four groups according to their sensitivity based on the means and the standard deviations of their total scores on seven main economic considerations. Nine fish species such as squids, hair tails, shellfishes, and crabs belonged to the hyper-sensitive group, whereas 15 fishes such as eels, sea breams, and sea weeds belonged to the sensitive group. Twelve species including common sea basses, cods, and abalones were among the less-sensitive group, and 11 species including skate rays and mud fishes comprised the non-sensitive group.

Performance Analysis of Transmit Weights Optimization for Cooperative Communications in Wireless Networks (무선네트워크의 협력통신을 위한 전송 무게(Transmit Weight) 최적화를 위한 연구)

  • Kong, Hyung-Yun;Ho, Van Khuong
    • The KIPS Transactions:PartC
    • /
    • v.12C no.7 s.103
    • /
    • pp.1025-1030
    • /
    • 2005
  • Cooperative communications among users in multiple access wireless environments is an efficient way to obtain the powerful benefits of multi-antenna systems without the demand for physical arrays. This paper proposes a solution to optimize the weights of partnering users' signals for the minimum error probability at the output of maximum likelihood (ML) detector under the transmit power constraints by taking advantage of channel state information (CSI) feedback from the receiver to the transmitter. Simulation programs are also established to evaluate the performance of the system under flat Rayleigh fading channel plus AWGN (Additive White Gaussian Noise).

Determining the Location of Urban Planning Measures for Preventing Debris-Flow Risks: Based on the MCDM Method (MCDM 기법을 이용한 도심지 토사재해 예방을 위한 도시계획적 대책 위치 결정방법 제안)

  • Moon, Yonghee;Lee, Sangeun;Kim, Soyoon;Kim, Myoungsoo
    • Journal of the Korean Society of Safety
    • /
    • v.32 no.5
    • /
    • pp.103-114
    • /
    • 2017
  • The landslide disaster damage has been increased by mountain development, leading to construction of educational facilities, medical facilities, petty industrial facilities, and large housing complexes. Therefore, effective regulation is required as an effort in urban planning solutions. For suggesting specific mitigation strategies on urban landslide, this study aims to define evaluation criteria for urban planning management of debris-flow disaster. AHP (Analytic Hierarchy Process), one of the multiple criterion decision making methods, was utilized in this study. This study makes use of 16 sub-criteria under the framework of hazard, exposure, and vulnerability, and well-planned expert survey measures their weights. The weights are also applied to evaluate each grid in urban space (min $10{\times}10m$) and classify it with red, orange, yellow, or green grade so that areas at higher risk are clearly identified. This study concludes that the suggested method is useful to support a strategies for urban planning management of debris-flow disaster, particularly in a GIS base.

POSITIVE SOLUTION FOR A CLASS OF NONLOCAL ELLIPTIC SYSTEM WITH MULTIPLE PARAMETERS AND SINGULAR WEIGHTS

  • AFROUZI, G.A.;ZAHMATKESH, H.
    • Journal of applied mathematics & informatics
    • /
    • v.35 no.1_2
    • /
    • pp.121-130
    • /
    • 2017
  • This study is concerned with the existence of positive solution for the following nonlinear elliptic system $$\{-M_1(\int_{\Omega}{\mid}x{\mid}^{-ap}{\mid}{\nabla}u{\mid}^pdx)div({\mid}x{\mid}^{-ap}{\mid}{\nabla}u{\mid}^{p-2}{\nabla}u)\\{\hfill{120}}={\mid}x{\mid}^{-(a+1)p+c_1}\({\alpha}_1A_1(x)f(v)+{\beta}_1B_1(x)h(u)\),\;x{\in}{\Omega},\\-M_2(\int_{\Omega}{\mid}x{\mid}^{-bq}{\mid}{\nabla}v{\mid}^qdx)div({\mid}x{\mid}^{-bq}{\mid}{\nabla}v{\mid}^{q-2}{\nabla}v)\\{\hfill{120}}={\mid}x{\mid}^{-(b+1)q+c_2}\({\alpha}_2A_2(x)g(u)+{\beta}_2B_2(x)k(v)\),\;x{\in}{\Omega},\\{u=v=0,\;x{\in}{\partial}{\Omega},$$ where ${\Omega}$ is a bounded smooth domain of ${\mathbb{R}}^N$ with $0{\in}{\Omega}$, 1 < p, q < N, $0{\leq}a$ < $\frac{N-p}{p}$, $0{\leq}b$ < $\frac{N-q}{q}$ and ${\alpha}_i,{\beta}_i,c_i$ are positive parameters. Here $M_i,A_i,B_i,f,g,h,k$ are continuous functions and we discuss the existence of positive solution when they satisfy certain additional conditions. Our approach is based on the sub and super solutions method.

An Integrated Approach to Measuring Supply Chain Performance

  • Theeranuphattana, Adisak;Tang, John C.S.;Khang, Do Ba
    • Industrial Engineering and Management Systems
    • /
    • v.11 no.1
    • /
    • pp.54-69
    • /
    • 2012
  • Chan and Qi (SCM 8/3 (2003) 209) developed an innovative measurement method that aggregates performance measures in a supply chain into an overall performance index. The method is useful and makes a significant contribution to supply chain management. Nevertheless, it can be cumbersome in computation due to its highly complex algorithmic fuzzy model. In aggregating the performance information, weights used by Chan and Qi-which aim to address the imprecision of human judgments-are incompatible with weights in additive models. Furthermore, the default assumption of linearity of its scoring procedure could lead to an inaccurate assessment of the overall performance. This paper addresses these limitations by developing an alternative measurement that takes care of the above. This research integrates three different approaches to multiple criteria decision analysis (MCDA)-the multiattribute value theory (MAVT), the swing weighting method and the eigenvector procedure-to develop a comprehensive assessment of supply chain performance. One case study is presented to demonstrate the measurement of the proposed method. The performance model used in the case study relies on the Supply Chain Operations Reference (SCOR) model level 1. With this measurement method, supply chain managers can easily benchmark the performance of the whole system, and then analyze the effectiveness and efficiency of the supply chain.

DLDW: Deep Learning and Dynamic Weighing-based Method for Predicting COVID-19 Cases in Saudi Arabia

  • Albeshri, Aiiad
    • International Journal of Computer Science & Network Security
    • /
    • v.21 no.9
    • /
    • pp.212-222
    • /
    • 2021
  • Multiple waves of COVID-19 highlighted one crucial aspect of this pandemic worldwide that factors affecting the spread of COVID-19 infection are evolving based on various regional and local practices and events. The introduction of vaccines since early 2021 is expected to significantly control and reduce the cases. However, virus mutations and its new variant has challenged these expectations. Several countries, which contained the COVID-19 pandemic successfully in the first wave, failed to repeat the same in the second and third waves. This work focuses on COVID-19 pandemic control and management in Saudi Arabia. This work aims to predict new cases using deep learning using various important factors. The proposed method is called Deep Learning and Dynamic Weighing-based (DLDW) COVID-19 cases prediction method. Special consideration has been given to the evolving factors that are responsible for recent surges in the pandemic. For this purpose, two weights are assigned to data instance which are based on feature importance and dynamic weight-based time. Older data is given fewer weights and vice-versa. Feature selection identifies the factors affecting the rate of new cases evolved over the period. The DLDW method produced 80.39% prediction accuracy, 6.54%, 9.15%, and 7.19% higher than the three other classifiers, Deep learning (DL), Random Forest (RF), and Gradient Boosting Machine (GBM). Further in Saudi Arabia, our study implicitly concluded that lockdowns, vaccination, and self-aware restricted mobility of residents are effective tools in controlling and managing the COVID-19 pandemic.

Improved Estimation Method for the Capacitor Voltage in Modular Multilevel Converters Using Distributed Neural Network Observer

  • Mehdi Syed Musadiq;Dong-Myung Lee
    • Journal of IKEEE
    • /
    • v.27 no.4
    • /
    • pp.430-438
    • /
    • 2023
  • The Modular Multilevel Converter (MMC) has emerged as a key component in HVDC systems due to its ability to efficiently transmit large amounts of power over long distances. In such systems, accurate estimation of the MMC capacitor voltage is of utmost importance for ensuring optimal system performance, stability, and reliability. Traditional methods for voltage estimation may face limitations in accuracy and robustness, prompting the need for innovative approaches. In this paper, we propose a novel distributed neural network observer specifically designed for MMC capacitor voltage estimation. Our observer harnesses the power of a multi-layer neural network architecture, which enables the observer to learn and adapt to the complex dynamics of the MMC system. By utilizing a distributed approach, we deploy multiple observers, each with its own set of neural network layers, to collectively estimate the capacitor voltage. This distributed configuration enhances the accuracy and robustness of the voltage estimation process. A crucial aspect of our observer's performance lies in the meticulous initialization of random weights within the neural network. This initialization process ensures that the observer starts with a solid foundation for efficient learning and accurate voltage estimation. The observer iteratively updates its weights based on the observed voltage and current values, continuously improving its estimation accuracy over time. The validity of proposed algorithm is verified by the result of estimated voltage at each observer in capacitor of MMC.