检索结果-内蒙古大学图书馆

Emerging Systems and Intelligent Computing (ESIC), International conference on

作者： Bishnu Pada Saha Satya Ranjan Dash Mainak Bandyopadhyay Paola Barra Vinh Truong Hoang School of Computer Engineering KIIT D.U. Bhubaneswar India School of Computer Applications KIIT D.U. Bhubaneswar India Department of Science and Technology University of Naples “Parthenope” Italy Faculty of Information Technology HCMC Open University Vietnam

ISBN: (数字)9798331522100

ISBN: (纸本)9798331522117

Attention mechanisms in vision applications of the computer world dynamically adjust feature weights to amplify the model’s capability to perceive information. This adaptive recalibration improves the overall effectiveness of the model. Spatial attention has seen widespread use to boost the efficiency of convolutional neural networks by enhancing their ability to model spatial information. This approach significantly improves CNNs’ spatial representation capabilities. Along this line of research, a novel binarized-attention module is drafted by taking spatial attention into consideration. Binary neural networks, which significantly reduce storage and computational requirements, are a promising approach for deploying deep models on resource-constrained devices. This technique offers efficiency in scenarios with limited resources. In this paper, we hereby implemented our algorithm of binarizing the spatial attention module and obtained diverse findings, demonstrating significant improvements in accuracies with fewer classes and degradation of accuracies with higher classes. Various experimental findings reveal that our technique delivers better accuracy with dataset having 10 classes but low accuracy with the one having 100 classes. Additionally, architectural optimizations, including the adoption of deeper networks like ResNet50, contribute to better handling of complex data patterns. While binarization offers computational efficiency, it often results in a trade-off with accuracy. However, selective binarization of specific network layers can mitigate this issue, maintaining a balance between performance and resource utilization. This study provides insights into the synergistic effects of these techniques, offering a pathway for developing efficient and accurate deep learning models suitable for deployment in resource-constrained environments.

关键词： Deep learning Adaptation models visualization Attention mechanisms Accuracy Computational modeling Convolutional neural networks Resource management optimization Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Classification and Prediction of Spinal Tuberculosis Disease Using optimization of Convolution Neural Network Using Spatial and Temporal Constraints 1st

Classification and Prediction of Spinal Tuberculosis Disease...

引用

1st International conference on Artificial Intelligence and its Applications, ICAIA 2023

作者： Askarali, K.T. Fredrik, E. J. Thomson Department of Computer Science Karpagam Academy of Higher Education Coimbatore India Department of Computer Technology Karpagam Academy of Higher Education Coimbatore India Department of Computer Science Malabar College of Advanced Studies Vengara India

ISBN: (纸本)9783031843969

Spinal Tuberculosis disease classification and prediction is avital task for identification of tuberculosis on bone as it leads to kyphosis, gibbus formation and osteomyelitis events. Those events have been classified into types and its stages on basis of various spatial and temporal properties of image morphology of the vertebral disk patterns in the symptomatic sites of the lumbar spine, thoracic spine obtained using MRI based imaging system. MRI imaging offers the excellent visualization of the vertebral disk patterns at the symptomatic sites. MRI images contain some sounds such as Gaussian noise, poisson noise, Impulse Noise and Quantization noises. Further the image contains complex appearance of the disk which makes difficult discriminate the stages and types of Spinal Tuberculosis Events. In order to manage those challenges, machine learning model using the unsupervised learning techniques becomes complex and time consuming. However deep learning architecture effectively discriminate the complex morphology structures of the vertebral disk patterns into spinal tuberculosis disease class events. In this paper, a Spatial and temporal constrained convolution Neural network (SPCCNN) is designed to classify and detect spinal tuberculosis into classes according to its severeness in levels. Initially Image noise eliminated through Wavelet Threshold based Image Denoising technique. Principle Component analysis has been employed to extract the feature of the preprocessed image. As samples of the image is enhanced using image augmentation. Image augmentation is an auxiliary convolution Neural network for increasing the size of the training set without obtaining images. A Convolution Neural network processes the features extracted by Principle component analysis at different layers of the network. The convolution layer and activation layer in particular are enhanced using spatial and temporal constraints on hyper parameters. Output layer produces the disease classes of t

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

High-Quality Ultra-Compact Grid Layout of Grouped Networks

引用

IEEE TRANSACTIONS ON visualization AND COMPUTER GRAPHICS 2016年第1期22卷 339-348页

作者： Yoghourdjian, Vahan Dwyer, Tim Gange, Graeme Kieffer, Steve Klein, Karsten Marriott, Kim Monash Univ Clayton Vic 3800 Australia Univ Melbourne Melbourne Vic 3010 Australia

Prior research into network layout has focused on fast heuristic techniques for layout of large networks, or complex multi-stage pipelines for higher quality layout of small graphs. Improvements to these pipeline techniques, especially for orthogonal-style layout, are difficult and practical results have been slight in recent years. Yet, as discussed in this paper, there remain significant issues in the quality of the layouts produced by these techniques, even for quite small networks. This is especially true when layout with additional grouping constraints is required. The first contribution of this paper is to investigate an ultra-compact, grid-like network layout aesthetic that is motivated by the grid arrangements that are used almost universally by designers in typographical layout. Since the time when these heuristic and pipeline-based graph-layout methods were conceived, generic technologies (MIP, CP and SAT) for solving combinatorial and mixed-integer optimization problems have improved massively. The second contribution of this paper is to reassess whether these techniques can be used for high-quality layout of small graphs. While they are fast enough for graphs of up to 50 nodes we found these methods do not scale up. Our third contribution is a large-neighborhood search meta-heuristic approach that is scalable to larger networks.

关键词： Network visualization graph drawing power graph optimization large-neighborhood search

来源：评论

学校读者我要写书评

暂无评论

Adaptable Radial Axes Plots for Improved Multivariate Data visualization

引用

COMPUTER GRAPHICS FORUM 2017年第3期36卷 389-399页

作者： Rubio-Sanchez, M. Sanchez, A. Lehmann, D. J. Univ Rey Juan Carlos Madrid Spain Ctr Computat Simulat Madrid Spain Univ Magdeburg Magdeburg Germany

Radial axes plots are multivariate visualization techniques that extend scatterplots in order to represent high-dimensional data as points on an observable display. Well-known methods include star coordinates or principal component biplots, which represent data attributes as vectors that define axes, and produce linear dimensionality reduction mappings. In this paper we propose a hybrid approach that bridges the gap between star coordinates and principal component biplots, which we denominate adaptable radial axes plots. It is based on solving convex optimization problems where users can: (a) update the axis vectors interactively, as in star coordinates, while producing mappings that enable to estimate attribute values optimally through labeled axes, similarly to principal component biplots;(b) use different norms in order to explore additional nonlinear mappings of the data;and (c) include weights and constraints in the optimization problems for sorting the data along one axis. The result is a flexible technique that complements, extends, and enhances current radial methods for data analysis.

关键词： Categories and Subject Descriptors (according to ACM CCS) [Human-centered computing]: visualization—visualization techniques [Probability and statistics]: Statistical paradigms—Statistical graphics [Human-centered computing]: visualization—visualization theory concepts and paradigms [Probability and statistics]: Statistical paradigms—Exploratory data analysis

来源：评论

学校读者我要写书评

暂无评论

Nonlinear programming and scientific computing visualization in the optimization design of electron optical system

引用

NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT 1999年第1-2期427卷 321-328页

作者： Gu, CX Liao, GY Jiang, HX Li, J Shan, LY Fudan Univ Dept Mat Sci Shanghai 200433 Peoples R China

In part I of this paper, five different nonlinear programming methods, improved Powell method, axis directional search method with fixed step, Hook-Jeeves method, axis directional search method with variable step and Rosenbrock method, are discussed for the optimization design of electron optical systems. Major aspects of these techniques in their applications are systematically studied, which include convergence rate, optimal objective function value and global convergence. The calculation results indicate that Rosenbrock method should be recommended in electron optical system optimization. In part II of this paper, an efficient method, reverse-match interpolation method, is presented for the visualized optimization design of electron optical systems, which is a new development of CAD in electron optics. This method can produce the density distribution of electron beam from electron trajectories worked out in the optimization search process. It makes real-time control system possible. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： electron optical system optimization design direct search optimization method visualization in scientific computing

来源：评论

学校读者我要写书评

暂无评论

Model-driven visualizations of constraint-based local search

引用

CONSTRAINTS 2009年第3期14卷 294-324页

作者： Dooms, Gregoire Van Hentenryck, Pascal Michel, Laurent Brown Univ Providence RI 02912 USA Univ Connecticut Storrs CT 06269 USA

visualization is often invaluable to understand the behavior of optimization algorithms, identify their bottlenecks or pathological behaviors, and suggest remedial techniques. Yet developing visualizations is often a tedious activity requiring significant time and expertise. This paper presents a framework for the visualization of constraint-based local search (CBLS) algorithms. Given a high-level model and a declarative visualization specification, the CBLS visualizer systematically produces animations to visualize constraints and objectives, violations, and conflicts, as well as the temporal behavior of these measures. The visualization specification is declarative and typically composed of a triple (what,where,how) indicating what to display, where, and with which graphical objects. The visualizer architecture is compositional and extensible. It provides building blocks which can be assembled freely by the user and focuses almost exclusively on static aspects, the dynamic aspects being automated by the use of invariants. The paper highlights various functionalities of the visualizer and describes a blueprint for its implementation.

关键词： Model-driven visualization Constraint-based local search optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Uniform Manifold Approximation with Two-phase optimization

Uniform Manifold Approximation with Two-phase Optimization

引用

IEEE visualization conference (VIS)

作者： Jeon, Hyeon Ko, Hyung-Kwon Lee, Soohyun Jo, Jaemin Seo, Jinwook Seoul Natl Univ Seoul South Korea NAVER Webtoon Corp Seongnam Si South Korea Sungkyunkwan Univ Seoul South Korea

ISBN: (数字)9781665488129

ISBN: (纸本)9781665488129

We introduce Uniform Manifold Approximation with Two-phase optimization (UMATO), a dimensionality reduction (DR) technique that improves UMAP to capture the global structure of high-dimensional data more accurately. In UMATO, optimization is divided into two phases so that the resulting embeddings can depict the global structure reliably while preserving the local structure with sufficient accuracy. In the first phase, hub points are identified and projected to construct a skeletal layout for the global structure. In the second phase, the remaining points are added to the embedding preserving the regional characteristics of local areas. Through quantitative experiments, we found that UMATO (1) outperformed widely used DR techniques in preserving the global structure while (2) producing competitive accuracy in representing the local structure. We also verified that UMATO is preferable in terms of robustness over diverse initialization methods, numbers of epochs, and subsampling techniques.

关键词： Human-centered computing visualization visualization techniques Computing methodologies Machine learning Machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

A new graphical visualization of n-dimensional Pareto front for decision-making in multiobjective optimization

引用

INFORMATION SCIENCES 2008年第20期178卷 3908-3924页

作者： Blasco, X. Herrero, J. M. Sanchis, J. Martinez, M. Univ Politecn Valencia Predict Control & Heurist Optimizat Grp Dept Syst Engn & Control Valencia 46022 Spain

New challenges in engineering design lead to multiobjective (multicriteria) problems. In this context, the Pareto front supplies a set of solutions where the designer (decision-maker) has to look for the best choice according to his preferences. visualization techniques often play a key role in helping decision-makers, but they have important restrictions for more than two-dimensional Pareto fronts. In this work, a new graphical representation, called Level Diagrams, for n-dimensional Pareto front analysis is proposed. Level Diagrams consists of representing each objective and design parameter on separate diagrams. This new technique is based on two key points: classification of Pareto front points according to their proximity to ideal points measured with a specific norm of normalized objectives (several norms can be used);and synchronization of objective and parameter diagrams. Some of the new possibilities for analyzing Pareto fronts are shown. Additionally, in order to introduce designer preferences, Level Diagrams can be coloured, so establishing a visual representation of preferences that can help the decision-maker. Finally, an example of a robust control design is presented - a benchmark proposed at the American Control conference. This design is set as a six-dimensional multiobjective problem. (c) 2008 Elsevier Inc. All rights reserved.

关键词： multiobjective optimization multidimensional visualization decision-making tools high dimensional Pareto front

来源：评论

学校读者我要写书评

暂无评论

Characterization and optimization of the flow pattern inside a diaphragm blood pump based on flow visualization techniques

引用

ASAIO JOURNAL 1998年第5期44卷 M714-M718页

作者： Nakata, M Masuzawa, T Tatsumi, E Taenaka, Y Nishimura, T Tsukiya, T Takano, H Tsuchimoto, K Ohba, K Ibaraki Univ Natl Cardiovasc Ctr Res Inst Hitachi Ibaraki Japan AISIN Cosmos Res & Dev Co Ltd Kariva Japan Kansai Univ Osuita Japan

We applied two different flow visualization techniques to obtain detailed information on the inside flow of the diaphragm blood pump of our electrohydraulic total artificial heart system to determine the optimum washout effect that would result in better antithrombogenicity. Major orifice directions of the inflow and outflow Bjork-Shiley valves of the left blood pump were independently changed to create 17 varied patterns. The character and velocity of the main flow at the diaphragm-housing junction were acquired using a laser light sheet method with polyethylene tracers. Wall shear flow, a major factor governing washout in the blood pump, was estimated by a newly developed paint erosion method. In this method, quantitative evaluation for an index of washout effect was made by calculating the residual ratio of the paint on the blood pump inner surface at 30 sec of pumping. When a single circular flow was consistently observed by the laser light sheet method, the paint residual ratio become low, indicating washout was relatively good. At the lowest paint residual ratio, the center of the circular flow observed by the laser light sheet method was located at the geometric center of the blood chamber. In conclusion, the flow pattern inside the blood pump could be characterized by combined use of these two flow visualization techniques, and the significant role of circular flow in better washout was clarified.

关键词： Heart Artificial

来源：评论

学校读者我要写书评

暂无评论

Interactive visualization of Dynamic and High-Dimensional Particle Swarm Behavior

Interactive Visualization of Dynamic and High-Dimensional Pa...

引用

IEEE International conference on Systems, Man, and Cybernetics (SMC)

作者： Wachowiak, Mark P. Sarlo, Bryan B. Nipissing Univ Dept Math & Comp Sci North Bay ON Canada

ISBN: (纸本)9781479906529

Particle swarm optimization (PSO) is a robust and popular stochastic population-based global optimization method that simulates social behavior among independent agents (particles). PSO is increasingly used to solve difficult high-dimensional and dynamic problems, where the global optima change over time. To better address the challenges inherent in these problems, interactive visualization is employed to study the behavior of these agents. In this paper, PSO variants are used to optimize high-dimensional and dynamic non-convex cost functions. Dimension reduction allows the application of state-of-the- art interactive scientific visualization techniques to study the behaviors and dynamic trends of the swarms, and to uncover patterns and algorithm mechanics. Problems in the search and weaknesses in the algorithms can be more easily identified, thereby facilitating enhancements for domain-specific problems. Results suggest that interactive visualization aids understanding of high-dimensional socially-based modeling.

关键词： visualization global optimization particle swarm dynamic optimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：