检索结果-内蒙古大学图书馆

An improved parallelism scheme for deterministic discrete ordinates transport

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 2018年第4期32卷 555-569页

作者： Deakin, Tom McIntosh-Smith, Simon Martineau, Matt Gaudin, Wayne Univ Bristol Dept Comp Sci Bristol Avon England UK Atom Weap Estab High Performance Comp Aldermaston England

In this paper we demonstrate techniques for increasing the node-level parallelism of a deterministic discrete ordinates neutral particle transport algorithm on a structured mesh to exploit many-core technologies. Transport calculations form a large part of the computational workload of physical simulations and so good performance is vital for the simulations to complete in reasonable time. We will demonstrate our approach utilizing the SNAP mini-app, which gives a simplified implementation of the full transport algorithm but remains similar enough to the real algorithm to act as a useful proxy for research purposes. We present an OpenCL implementation of our improved algorithm which achieves a speedup of up to 2.5 x on a many-core GPGPU device compared to a state-of-the-art multi-core node for the transport sweep, and up to 4 x compared to the multi-core CPUs in the largest GPU enabled supercomputer;the first time this scale of speedup has been achieved for algorithms of this class. We then discuss ways to express our scheme in OpenMP 4.0 and demonstrate the performance on an Intel Knights Corner Xeon Phi compared to the original scheme.

关键词： computer programming MOBILE apps computer algorithms CENTRAL processing units (computers) GRAPHICS processing units (computers)

来源：评论

学校读者我要写书评

暂无评论

ENERGY-AWARE NAVIGATION IN LARGE-SCALE EVACUATION USING G-NETWORKS

引用

PROBABILITY IN THE ENGINEERING AND INFORMATIONAL SCIENCES 2018年第3期32卷 340-352页

作者： Bi, Huibo Abdelrahman, Omer H. Imperial Coll Dept Elect & Elect Engn Intelligent Syst & Networks Grp London SW7 2AZ England

Previous studies on emergency management of large-scale urban networks have commonly concentrated on system development to off-load intensive computations to remote cloud servers or improving communication quality during a disaster and ignored the effect of energy consumption of vehicles, which can play a vital role in large-scale evacuation owing to the disruptions in energy supply. Hence, in this paper we propose a cloud-enabled navigation system to direct vehicles to safe areas in the aftermath of a disaster in an energy and time efficient fashion. A G-network model is employed to mimic the behaviors and interactions between individual vehicles and the navigation system, and analyze the effect of re-routing decisions toward the vehicles. A gradient descent optimization algorithm is used to gradually reduce the evacuation time and fuel consumption of vehicles by optimizing the probabilistic choices of linked road segments at each intersection. The re-routing decisions arrive at the intersections periodically and will expire after a short period. When a vehicle reaches an intersection, if the latest re-routing decision has not expired, the vehicle will follow this advice, otherwise, the vehicle will stick to the shortest path to its destination. The experimental results indicate that the proposed algorithm can reduce the evacuation time and the overall fuel utilization especially when the number of evacuated vehicles is large.

关键词： EMERGENCY management ENERGY consumption LARGE scale systems ROUTING (computer network management) MATHEMATICAL optimization computer algorithms

来源：评论

学校读者我要写书评

暂无评论

A physical-statistical model for the growth process of carbon nanotubes

引用

APPLIED PHYSICS A-MATERIALS SCIENCE & PROCESSING 2018年第11期124卷 1-10页

作者： Fan, Mengqi Wu, Su Xu, Chi Tsinghua Univ Dept Ind Engn Beijing 100084 Peoples R China

In the growth process of carbon nanotube (CNT) arrays, the height and flatness of the array must be controlled. In general, a profile control algorithm can be used to characterize the morphology of the CNT array through considering the parameters of the profile function as control objective. This paper examines the performance of the present model using real data and finds the defect of losing correlation information in the data. To solve this limitation, the semivariogram borrowed from geostatistics is introduced to add a spatial correlation. Furthermore, we drive the growth height formula based on a physical mechanism and then propose a physical-statistical model that contains a spatial correlation. Simulation and validation are provided to verify the performance of the new model.

关键词： observational data arrays Carbon Nanotubes computer algorithms flatness geostatistics

来源：评论

学校读者我要写书评

暂无评论

A Strong Tracking SLAM Algorithm Based on the Suboptimal Fading Factor

引用

JOURNAL OF SENSORS 2018年第5-3期2018卷 1-14页

作者： Liang, Yunpei Dai, Jiahui Wang, Kequan Li, Xiaobo Xu, Pengcheng Chongqing Univ State Key Lab Coal Mine Disaster Dynam & Control Chongqing 400044 Peoples R China Chongqing Res Inst China Coal Technol Engn Grp Chongqing 400037 Peoples R China

This paper proposes an innovative simultaneous localization and mapping (SLAM) algorithm which combines a strong tracking filter (STF), an unscented Kalman filter (UKF), and a particle filter (PF) to deal with the low accuracy of unscented FastSLAM (UFastSLAM). UFastSLAM lacks the capacity for online self-adaptive adjustment, and it is easily influenced by uncertain noise. The new algorithm updates each Sigma point in UFastSLAM by an adaptive algorithm and obtains optimized filter gain by the STF adjustment factor. It restrains the influence of uncertain noise and initial selection. Therefore, the state estimation would converge to the true value rapidly and the accuracy of system state estimation would be improved eventually. The results of simulations and practical tests show that strong tracking unscented FastSLAM (STUFastSLAM) has a significant improvement in accuracy and robustness.

关键词： SLAM (Robotics) Adaptive control systems Kalman filtering Tracking filters Monte Carlo method computer algorithms

来源：评论

学校读者我要写书评

暂无评论

A hybrid evolutionary preprocessing method for imbalanced datasets

引用

INFORMATION SCIENCES 2018年 454卷 161-177页

作者： Wong, Ginny Y. Leung, Frank H. F. Ling, Sai-Ho Hong Kong Polytech Univ Dept Elect & Informat Engn Ctr Signal Proc Hong Kong Hong Kong Peoples R China Univ Technol Sydney Fac Engn & Informat Technol Ctr Hlth Technol Sydney NSW Australia

Imbalanced datasets are commonly encountered in real-world classification problems. Many machine learning algorithms are originally designed for well-balanced datasets, therefore re-sampling has become an important step to pre-process imbalanced data. This aims to balance the datasets by increasing the samples of the smaller class or decreasing the samples of the larger class, which are known as over-sampling and under-sampling, respectively. In this paper, a sampling strategy that is based on both over-sampling and under-sampling is proposed, in which the new samples of the smaller class are created based on fuzzy logic. Improvement of the datasets is done by the evolutionary computational method of Cross-generational elitist selection, Heterogeneous recombination and Cataclysmic mutation (CHC) that under-samples both the minority and majority samples. Consequently, a hybrid preprocessing method is proposed to re-sample imbalanced datasets. The evaluation is done by applying the Support Vector Machine (SVM), C4.5 decision tree and nearest neighbor rule to train a classification model from the re-sampled training sets. From the experimental results, it can be seen that our proposed method improves both the F-measure and AUC. The over-sampling rate and complexity of the classification model are also compared. Our proposed method is found to be superior to all other methods under comparison and it is more robust in different classifiers. (C) 2018 Published by Elsevier Inc.

关键词： HYBRID systems EVOLUTIONARY theories MACHINE learning computer algorithms SAMPLING (Statistics)

来源：评论

学校读者我要写书评

暂无评论

Crowd evacuation simulation approach based on navigation knowledge and two-layer control mechanism

引用

INFORMATION SCIENCES 2018年 436卷 247-267页

作者： Liu, Hong Liu, Baoxi Zhang, Hao Li, Liang Qin, Xin Zhang, Guijuan Shandong Normal Univ Sch Informat Sci & Engn 88 East Wenhua Rd Jinan Shandong Peoples R China Shandong Prov Key Lab Novel Distributed Comp Soft Jinan Shandong Peoples R China

This paper presents a crowd evacuation simulation approach that is based on navigation knowledge and two-layer control mechanism. In this approach, using the multi-population cultural algorithm framework, the control mechanism of the crowd evacuation simulation is divided into two parts, namely, the belief and population spaces. The population space is divided into groups (sub-populations), and a leader is selected in each group according to a fitness value. The belief space comprises multiple agents and a knowledge base. Each navigation agent corresponds to a group leader. A navigation agent obtains a leader's position through the acceptance function and later passes the information to the knowledge base. On the basis of the position, the obstacles, and the congestion situation provided by the navigation agent, the knowledge base management agent dynamically plans the path and provides the navigation agent the next position along the path. The navigation agent later passes the information to the leader through 'the affection function. The individuals in the group follow the leader through the social force model in moving to the location provided by the navigation agent. The entire process is repeated until the exit is reached. The path information that successfully reached the exit is recorded, and the knowledge base is updated. This method establishes the relationship between the population and the navigation agent with knowledge and transforms a blindly moving crowd into a guided evacuation as the mass evacuation simulation problem is decomposed into a sub-problem of moving blocks. This approach effectively solves the problem of microscopic models because each individual calculates the path and resolves the slow speed problem. The simulation results illustrate the effectiveness of this method. (C) 2018 Elsevier Inc. All rights reserved.

关键词： NAVIGATION CONTROL theory (Mathematics) computer algorithms DATA flow computing SIMULATION methods & models

来源：评论

学校读者我要写书评

暂无评论

Relational Neurogenesis for Lifelong Learning Agents

Relational Neurogenesis for Lifelong Learning Agents

引用

作者： Pandit, Tej Rochester Institute of Technology

学位级别：M.S.

Reinforcement learning systems have shown tremendous potential in being able to model meritorious behavior in virtual agents and robots. The ability to learn through continuous reinforcement and interaction with an environment negates the requirement of painstakingly curated datasets and hand crafted features. However, the ability to learn multiple tasks in a sequential manner, referred to as lifelong or continual learning, remains unresolved. The search for lifelong learning algorithms creates the foundation for this work. While there has been much research conducted in supervised learning domains under lifelong learning, the reinforced lifelong learning domain remains open for much exploration. Furthermore, current implementations either concentrate on preserving information in fixed capacity networks, or propose incrementally growing networks which randomly search through an unconstrained solution space. In order to develop a comprehensive lifelong learning algorithm, it seems essential to amalgamate these approaches into a condensed algorithm which can perform both neuroevolution and constrict network growth automatically. This thesis proposes a novel algorithm for continual learning using neurogenesis in reinforcement learning agents. It builds upon existing neuroevolutionary techniques, and incorporates several new mechanisms for limiting the memory resources while expanding neural network learning capacity. The algorithm is tested on a custom set of sequential virtual environments which emulate several meaningful scenarios for intellectually down-scaled species and autonomous robots. Additionally, a library for connecting an unconstrained range of machine learning tools, in a variety of programming languages to the Unity3D simulation engine for the development of future learning algorithms and environments, is also proposed.

关键词： Machine learning computer algorithms Neural networks (computer science)

来源：评论

学校读者我要写书评

暂无评论

A nomenclature algorithm for a potentially global soil taxonomy

引用

GEODERMA 2018年 322卷 56-70页

作者： Hughes, Philip McBratney, Alex B. Huang, Jingyi Minasny, Budiman Micheli, Erika Hempel, Jonathan Univ Sydney Sydney Inst Agr Eveleigh NSW 2015 Australia Univ Sydney Sch Life & Environm Sci Eveleigh NSW 2015 Australia UNSW Sch Biol Earth & Environm Sci Sydney NSW 2052 Australia Szent Istvan Univ Dept Soil Sci & Agr Chem Godollo Hungary US Dept Agr Nat Resources Conservat Serv Natl Soil Survey Ctr Lincoln NE 68508 USA

Previous work has been put into the creation of a global soil taxonomy using a harmonised dataset of 23 soil properties at 18 depth intervals. The taxonomy consisted of selected soil taxa from the US Soil Taxonomy, World Reference Base for Soil Resources, the Australian Soil Classification, and the New Zealand Soil Classification. In this paper, a nomenclature algorithm was proposed for this established comprehensive taxonomy. Firstly, a Ward dendrogram was calculated from a weighted distance matrix determined from principal components of the taxa. This dendrogram was then cut at three levels, creating 15 groups, 86 subgroups, and 493 sub-subgroups at tiers 1, 2 and 3, respectively. A sequence of consonants was used to name the taxa at each tier alphabetically with "A" and "E" inserted between the consonants of tiers 2 and 3 and "OZEM" appended after the consonants of tier 1. In addition, a distance-based algorithm was used to allocate and name 10 unknown soil profiles to the comprehensive soil taxonomy. It was concluded that the nomenclature algorithm can be easily disaggregated by computer and can be used to understand the inter-relationships between soil profiles from different classification systems. In the future, there is a need to include other soil classification systems to the comprehensive system and assign different weights to the soil properties and depths used to construct the comprehensive soil classification system.

关键词： SOIL classification SOIL dynamics SOIL profiles computer algorithms PRINCIPAL components analysis

来源：评论

学校读者我要写书评

暂无评论

KeratoDetect: Keratoconus Detection Algorithm Using Convolutional Neural Networks

引用

COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2019年第1期2019卷 8162567-8162567页

作者： Lavric, Alexandru Valentin, Popa Stefan Cel Mare Univ Suceava Comp Elect & Automat Dept Suceava 720229 Romania

Keratoconus (KTC) is a noninflammatory disorder characterized by progressive thinning, corneal deformation, and scarring of the cornea., e pathological mechanisms of this condition have been investigated for a long time. In recent years, this disease has come to the attention of many research centers because the number of people diagnosed with keratoconus is on the rise. In this context, solutions that facilitate both the diagnostic and treatment options are quickly needed., emain contribution of this paper is the implementation of an algorithm that is able to determine whether an eye is affected or not by keratoconus., e KeratoDetect algorithm analyzes the corneal topography of the eye using a convolutional neural network (CNN) that is able to extract and learn the features of a keratoconus eye., e results show that the KeratoDetect algorithm ensures a high level of performance, obtaining an accuracy of 99.33% on the data test set. KeratoDetect can assist the ophthalmologist in rapid screening of its patients, thus reducing diagnostic errors and facilitating treatment.

关键词： KERATOCONUS NEURAL networks (Neurobiology) DIAGNOSIS CORNEA computer algorithms

来源：评论

学校读者我要写书评

暂无评论

Is Your Algorithm Dangerous?

引用

IEEE TECHNOLOGY AND SOCIETY MAGAZINE 2018年第3期37卷 20-21页

作者： Moses, Lyria Bennett UNSW Law Allens Hub Technol Law & Innovat Sydney NSW 2052 Australia

Much has been written about artificial intelligence, both from the perspective of possibilities and opportunities as well as from the perspective of risks and limitations. Here I make a simple point - evaluating the appropriateness of an algorithm requires understanding the domain space in which it will operate. While data science enables one to transcend expertise in a particular domain, it nevertheless requires a deep familiarity with the question it is required to answer. Focusing on the answer rather than the question presents significant dangers. These are not necessarily physical hazards, but rather dangers to things like social norms, rule of law values, and the experience of equality. Deploying algorithms that do not avoid these dangers risks injustice in individual cases as well as generating longer term threats to fundamental social and democratic values. Consider the context of criminal justice. Risk assessment tools are increasingly used, particularly in the United States, to make decisions about bail, parole, and sentencing.

关键词： computer algorithms CRIMINAL justice system SOCIAL values ARTIFICIAL intelligence DATA science

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：