检索结果-内蒙古大学图书馆

Improving the prediction accuracy of cyanobacterial blooms using a hybrid model integrating dimensionality reduction and deep learning

引用

Water Biology and Security 2025年

作者： Jiwei Chen Zhongbao Hou Yunxi Li Jiexian Li Lin Li School of Naval Architecture and Ocean Engineering Huazhong University of Science and Technology Wuhan 430074 China Department of Public Health Fu Jen Catholic University Taiwan 242062 China Khoury College of Computer Sciences Northeastern University Vancouver V6B1Z3 Canada Institute of Hydrobiology Chinese Academy of Sciences Wuhan 430072 China

Cyanobacteria Harmful Blooms have negatively impacted human socio-economic activities and the natural ecological environment. Predicting cyanobacterial blooms requires the application of advanced analytical models powered by artificial intelligence to accurately model bloom development. In this study, we developed a hybrid model that integrates the Temporal Convolution Network (TCN) with the Long Short-Term Memory Network (LSTM) to achieve highly accurate predictions of cyanobacterial bloom events. We used Principal Component Analysis (PCA) during the preprocessing phase to reduce data dimensionality of model input variables and enhance computational efficiency. We used a data set from the Darling River in Australia to conduct a comparative analysis with traditional deep learning models and comprehensively assess the proposed model performance. We also investigated the application of the correlation coefficient matrix method in variable selection. The results demonstrate that the proposed model outperforms conventional methods in predicting cyanobacterial blooms. When benchmarked against the Recurrent Neural Network (RNN) model, our model demonstrates a superior ability to capture the intricacies of time series data. Specifically, it achieves a 79.5 % reduction in Mean Square Error (MSE) and a 197 % increase in the Coefficient of Determination (R 2 ). Furthermore, comparative experiments were carried out across several prediction time steps, indicating that the model achieves optimal performance in minimizing both the MSE and Mean Absolute Error (MAE) when predicting one day ahead. Notably, the error distribution is most condensed at this time step, outperforming models with prediction horizons ranging from two to seven days. We also undertook a thorough analysis of the significance of each principal component in model predictions. The results highlight that principal component 5 and principal component 2 exhibit the most substantial contribution to the prediction

关键词： Principal component analysis Deep learning Cyanobacteria bloom Prediction Temporal convolution network Long short-term memory network

来源：评论

学校读者我要写书评

暂无评论

An extended analysis of memory hierarchies for efficient implementations of image processing applications

An extended analysis of memory hierarchies for efficient imp...

引用

作者： Hartmann, Christian Fey, Dietmar Chair of Computer Architecture University of Erlangen-Nuremberg Martensstr. 3 Erlangen91058 Germany

Through continued miniaturization of electronic devices embedded smart cameras are steadily becoming more and more important. The reduction of the camera size increases the spectrum of applications. In industrial applications the range of smart cameras spans from quality monitoring and position tracking to the calibration of production machines. In non-professional applications a distinct boom in action cameras combined with fused sensor information can be observed. However, all of these applications have a common bottleneck: the memory architecture. Most image processing applications are memory-bound tasks. Thus, the amount of time for transferring data with image processing applications decisively affects the application’s entire processing time. Different memory access patterns require different memory configurations and hierarchies. An insufficient match between the image processing application and the memory architecture leads to a poor performance in the image processing system. This can lead to longer processing times, and larger energy consumption rates. This work introduces new methods of classifying image processing applications by using their memory access pattern for mapping on memory architectures. Our work combines a simulation framework the heterogenous memory simulator with a analytical framework the memory analyzer to find bottlenecks inside the image processing application and aids in finding a suitable, application-specific memory configuration in terms of processing time and energy consumption. © 2017, Springer-Verlag GmbH Germany.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

IPAS: a design framework for analysis, synthesis and optimization of image processing applications for heterogenous computing architectures

IPAS: a design framework for analysis, synthesis and optimiz...

引用

作者： Hartmann, C. Häublein, K. Reichenbach, M. Fey, D. Chair of Computer Architecture University of Erlangen-Nuremberg Martensstr. 3 Erlangen91058 Germany

Recent trends in the image processing field have led to the use of more heterogeneous hardware architectures. The reason for this increase is that specialized cores, compared to standard CPUs, offer a more efficient way of achieving image processing applications. Specialized cores have less power, resource, and area consumption. On the other hand, designing such a heterogenous system with specialized cores is a challenging, error-prone and time-consuming task. Therefore, new frameworks are necessary for bringing an image processing application onto a given target platform by means of a tool chain. Some frameworks exist, but they do not address each need of a heterogeneous image processing application. Common weaknesses are (1) the low utilization of the image processing domain, (2) the inflexibility of the programming paradigms for different hardware architectures. Therefore, we define our own domain-specific design language called IPOL. To automate the derivation and optimization process, a synthesis tool named Image Processing architecture Synthesis was created. This tool will be the focus of this work. © 2016, Springer-Verlag Berlin Heidelberg.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

TDMP-Reliable Target Driven and Mobility Prediction based Routing Protocol in Complex VANET

arXiv

引用

arXiv 2020年

作者： Ye, Mao Guan, Lin Quddus, Mohammed Department of Computer Science Loughborough University LoughboroughLE11 3TU United Kingdom School of Architecture Building and Civil Engineering Loughborough University LoughboroughLE11 3TU United Kingdom

Vehicle-to-everything (V2X) communication in vehicular ad hoc network (VANET) has emerged as a crucial component in advanced Intelligent Transport System (ITS) for information transmission and vehicular communication. One of the vital research challenges in VANET is the design and implementation of novel network routing protocols which bring reliable end-to-end connectivity and efficient packet transmission to V2X communication. The organically changing nature of road traffic vehicles poses a significant threat to VANET with respect to the accuracy and reliability of packets delivery. Therefore, position-based routing protocols tend to be the predominant method in VANET as they overcome rapid changes in vehicle movements effectively. However, existing routing protocols have some limitations such as (i) inaccurate in high dynamic network topology, (ii) defective link-state estimation (iii) poor movement prediction in heterogeneous road layouts. Therefore, a novel target-driven and mobility prediction (TDMP) based routing protocol is developed in this paper for high-speed mobility and dynamic topology of vehicles, fluctuant traffic flow and diverse road layouts in VANET. To implement an effective routing protocol, TDMP primarily involves the destination target of a driver for the mobility prediction and Received Signal Strength Indicator (RSSI) for the inter-vehicular link-status estimation. Compared to existing geographic routing protocols which mainly greedily forward the packet to the next-hop based on its current position and partial road layout, the proposed TDMP is able to enhance the packet transmission with the consideration of the estimation of inter-vehicular link status, and the prediction of vehicle positions dynamically in fluctuant mobility and global road layout. Based on the extensive simulations carried out on operational road environments with varying configurations and complexity, the experimental results show better performance in terms of improvin

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

MTLSegFormer: Multi-task Learning with Transformers for Semantic Segmentation in Precision Agriculture

arXiv

引用

arXiv 2023年

作者： Goncalves, Diogo Nunes Junior, Jose Marcato Zamboni, Pedro Pistori, Hemerson Li, Jonathan Nogueira, Keiller Goncalves, Wesley Nunes Faculty of Computer Science Federal University of Mato Grosso do Sul Av. Costa e Silva MS Campo Grande79070-900 Brazil Faculty of Engineering Architecture and Urbanism and Geography Federal University of Mato Grosso do Sul Av. Costa e Silva MS Campo Grande79070-900 Brazil INOVISAO Dom Bosco Catholic University Avenida Tamandaré 6000 MS Campo Grande79117-900 Brazil Department of Geography and Environmental Management University of Waterloo WaterlooONN2L 3G1 Canada University of Stirling StirlingFK9 4LA United Kingdom

Multi-task learning has proven to be effective in improving the performance of correlated tasks. Most of the existing methods use a backbone to extract initial features with independent branches for each task, and the exchange of information between the branches usually occurs through the concatenation or sum of the feature maps of the branches. However, this type of information exchange does not directly consider the local characteristics of the image nor the level of importance or correlation between the tasks. In this paper, we propose a semantic segmentation method, MTLSegFormer, which combines multi-task learning and attention mechanisms. After the backbone feature extraction, two feature maps are learned for each task. The first map is proposed to learn features related to its task, while the second map is obtained by applying learned visual attention to locally re-weigh the feature maps of the other tasks. In this way, weights are assigned to local regions of the image of other tasks that have greater importance for the specific task. Finally, the two maps are combined and used to solve a task. We tested the performance in two challenging problems with correlated tasks and observed a significant improvement in accuracy, mainly in tasks with high dependence on the others. Copyright © 2023, The Authors. All rights reserved.

关键词： Precision agriculture

来源：评论

学校读者我要写书评

暂无评论

An Empirical Investigation of the Role of Pre-training in Lifelong Learning

arXiv

引用

arXiv 2021年

作者： Mehta, Sanket Vaibhav Patil, Darshan Chandar, Sarath Strubell, Emma School of Computer Science Carnegie Mellon University PittsburghPA15213 United States Mila Quebec AI Institute Université de Montréal MontrealQCH3T 1J4 Canada Mila Quebec AI Institute Canada CIFAR AI Chair École Polytechnique de Montréal MontrealQCH3T 1J4 Canada

The lifelong learning paradigm in machine learning is an attractive alternative to the more prominent isolated learning scheme not only due to its resemblance to biological learning but also its potential to reduce energy waste by obviating excessive model re-training. A key challenge to this paradigm is the phenomenon of catastrophic forgetting. With the increasing popularity and success of pre-trained models in machine learning, we pose the question: What role does pre-training play in lifelong learning, specifically with respect to catastrophic forgetting? We investigate existing methods in the context of large, pre-trained models and evaluate their performance on a variety of text and image classification tasks, including a large-scale study using a novel data set of 15 diverse NLP tasks. Across all settings, we observe that generic pre-training implicitly alleviates the effects of catastrophic forgetting when learning multiple tasks sequentially compared to randomly initialized models. We then further investigate why pre-training alleviates forgetting in this setting. We study this phenomenon by analyzing the loss landscape, finding that pre-trained weights appear to ease forgetting by leading to wider minima. Based on this insight, we propose jointly optimizing for current task loss and loss basin sharpness to explicitly encourage wider basins during sequential fine-tuning. We show that this optimization approach outperforms several state-of-the-art task-sequential continual learning algorithms across multiple settings, occasionally even without retaining a memory that scales in size with the number of tasks. © 2021, CC BY.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Pricing Strategy for Household Energy-Saving Option (Heso): A Novel Option-Based Intervention for Promoting Household Energy Efficiency

SSRN

引用

SSRN 2022年

作者： Xu, Qian Li, Shengping Shen, Liyin Chang, Ruidong Wang, Qian-Cheng Liu, Xuan Department of the Built Environment College of Design and Engineering National University of Singapore 4 Architecture Drive Singapore117566 Singapore Department of Infrastructure Engineering Faulty of Engineering and Information Technology The University of Melbourne VIC3010 Australia School of Spatial Planning and Design Zhejiang University City College Hangzhou China School of Management Science and Real Estate Chongqing University Chongqing China School of Architecture and Built Environment Faculty of Engineering Computer and Mathematical Sciences The University of Adelaide SA5005 Australia Department of Land Economy University of Cambridge CambridgeCB3 9EP United Kingdom Department of the Built Environment Eindhoven University of Technology Eindhoven5600MB Netherlands

Household energy-saving interventions are essential in reducing energy consumption in residential buildings. One of the effective interventions is Household Energy-Saving Option (HESO). However, the pricing strategy of this innovative intervention has not been explored. This study aims to establish a feasible valuation strategy for HESO to facilitate the application of market-based energy-saving interventions among households. This study first provides a pricing strategy of HESO from a market zero-profit perspective. It continues by developing a price cap and floor for HESO's market price. The theoretical pricing model was validated by a field experiment consisting of two-tiered HESO trading among 67 households in Singapore. The results revealed decisive implications for both environmental policymakers and HESO suppliers. Policymakers are suggested to launch HESO in countries or regions with stable and relatively high electricity prices and low discount rates. HESO suppliers may choose to adjust rewards and duration accordingly when the market price deviates from the zero-profit equilibrium price of HESO. The valuation strategy of HESO also helps quantitative estimation of households’ perceived value of HESO. © 2022, The Authors. All rights reserved.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Elastic waves in periodically anisotropic heterogeneous media: bridge the gap between rigorous and phenomenological approaches

引用

Journal of Physics: Conference Series 2024年第25期2647卷 252034-252034页

作者： I. Andrianov V. Danishevskyy J. Kaplunov Y. Kirichek Chair and Institute of General Mechanics RWTH Aachen University 18 Eilfschornstein-strasse Aachen 52062 Germany Prydniprovska State Academy of Civil Engineering and Architecture 24a Architect Oleg Petrov street Dnipro 49005 Ukraine School of Computer Science and Mathematics Keele University Keele ST55BG UK 4 Faculty of Civil Engineering and Geosciences Delft University of Technology 1 Stevinweg Delft 2628CN The Netherlands

Despite the growing capacity of computer codes, analytical solutions are still of great interest. As a rule, they are based on certain asymptotic approximations. In our work, we use a two-scale asymptotic procedure. Anti-plane shear waves in a layered medium are studied. To clarify the basics of the methodology, we restrict ourselves with a layered membrane. For long-wave case we obtained solutions for periodic and anti-periodic modes. We analyse them in the low- and high-contrast cases. The results obtained can be generalized for complex multiscale heterogeneous media and structures. They are useful for bridging the gap between mathematically rigorous and phenomenological approaches in dynamics of heterogeneous materials. They also can be implicated as benchmarks for numerical modelling.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Corrigendum to ‘Research on Lightweight Anomaly Detection of Multimedia Traffic in Edge Computing’ computer & Security, 111(2021) 102463

引用

computers & Security 2023年 125卷

作者： Xu Zhao Guangqiu Huang Jin Jiang Ling Gao Maozhen Li School of Management Xi'an University of Architecture & Technology Xi'an 710055 China School of Computer Science Xi'an Polytechnic University Xi'an 710048 China Department of Electronic and Electrical Engineering Brunel University London Uxbridge UB8 3PH UK School of Humanities Xi'an Polytechnic University Xi'an 710048 China

来源：评论

学校读者我要写书评

暂无评论

Reviewing two decades of Energy system analysis with bibliometrics

arXiv

引用

arXiv 2021年

作者： Dominković, Dominik Franjo Weinand, Jann Michael Scheller, Fabian D'Andrea, Matteo McKenna, Russell Department of Applied Mathematics and Computer Science Technical University of Denmark Matematiktorvet Kgs. Lyngby2800 Denmark Karlsruhe Institute of Technology Karlsruhe Germany Energy Systems Analysis Division of Sustainability Department of Technology Management and Economics Technical University of Denmark Matematiktorvet Kgs. Lyngby2800 Denmark Chair of Energy Transition School of Engineering University of Aberdeen King's College AberdeenAB24 3FX United Kingdom

The field of Energy System Analysis (ESA) has experienced exponential growth in the number of publications since at least the year 2000. This paper presents a comprehensive bibliometric analysis on ESA by employing different algorithms in Matlab and R. The focus of results is on quantitative indicators relating to number and type of publication outputs, collaboration links between institutions, authors and countries, and dynamic trends within the field. The five and twelve most productive countries have 50% and 80% of ESA publications respectively. The dominant institutions are even more concentrated within a small number of countries. A significant concentration of published papers within countries and institutions was also confirmed by analysing collaboration networks. These show dominant collaboration within the same university or at least the same country. There is also is a strong link among the most successful journals, authors and institutions. The Energy journal has had the most publications in the field, and its editor-in-chief Lund H is the author with most of the publications in the field, as well as the author with most of the highly cited publications in the field. In terms of the dynamics within the field in the past decade, recent years have seen a higher impact of topics related to flexibility and hybrid/integrated energy systems alongside a decline in individual technologies. This paper provides a holistic overview of two decades' research output and enables interested readers to obtain a comprehensive overview of the key trends in this active field. © 2021, CC BY-NC-ND.

关键词： Systems analysis

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：