检索结果-内蒙古大学图书馆

The impact of backbone selection in YOLOv8 models on brain tumor localization

Iran Journal of computer science 2025年 1-23页

作者： Ranjbarzadeh, Ramin Crane, Martin Bendechache, Malika School of Computing Faculty of Engineering and Computing Dublin City University Dublin Ireland ADAPT Research Centre School of Computer Science University of Galway Galway Ireland

This study investigates the utilization of the You Only Look Once (YOLOv8) deep learning framework for accurately identifying the location of brain tumors in medical imaging. We investigate the effects of model size and pretraining on the accuracy and computational efficiency of tumor detection by utilizing different setups of the YOLOv8 model. These setups include various configurations, ranging from very small to large, and can be pretrained on the COCO dataset or not. The experimental results, carried out on Google Colaboratory using NVIDIA Tesla T4 GPUs, show that pretrained models often achieve better performance by utilizing the extensive feature representations learned from the COCO dataset, resulting in increased precision in tumor location. For instance, the YOLOv8-XS model pretrained on COCO achieves an IoU of 0.278 and a Dice coefficient of 0.435, whereas its non-pretrained counterpart attains only 0.241 IoU and 0.388 Dice, indicating a 15% improvement in tumor localization accuracy. Similarly, pretrained YOLOv8-L achieves 0.269 IoU, outperforming standard object detection models such as Mask R-CNN (IoU: 0.212) and Faster R-CNN (IoU: 0.228). These results highlight the impact of pretraining on model performance, particularly for lightweight architectures, while also revealing diminishing returns for larger models. The research uncovers a subtle connection between the complexity of the model, pretraining, and the time required for training. It emphasizes the possible advantages and constraints of pretraining for various sizes of models. © The Author(s) 2025.

关键词： Brain tumor localization COCO dataset Deep learning Object detection Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Activation Control of Vision Models for Sustainable AI Systems

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第7期5卷 3470-3481页

作者： Burton-Barr, Jonathan Fernando, Basura Rajan, Deepu Institute of High-Performance Computing The Centre for Frontier AI Research Agency for Science Technology and Research 138632 Singapore Nanyang Technological University Nanyang School of Computer Science and Engineering 639798 Singapore

As artificial intelligence (AI) systems become more complex and widespread, they require significant computational power, increasing energy consumption. Addressing this challenge is essential for ensuring the long-term sustainability of AI technology. AI-on-AI control refers to a system with a set of AI functions controlled by an upper-level AI model. Previous work in AI-on-AI control focuses on boosting accuracy or expanding system capability by increasing overall system cost. Alternatively, we focus on applying AI-on-AI control to decrease system cost and increase the sustainability and viability of a system with multiple AI functions. Our supervised image classification evaluative controller (SICEC) is a cost-reduction oriented AI-on-AI controller that learns when vision models within an AI system should be activated based on input features. The function controller (FC) preprocesses an input and activates relevant functions, functions being distinct units of AI functionality within the system. Some functions have a set of same functional models (SFMs). These models take the same input and produce the same output but have architectural differences. We introduce a same functional controller to select a SFM using the FC's decision confidence. Results are promising, with a decrease of up to 48.9% in inference time, 67.8% in floating point operations (FLOPs), and 66.4% in energy usage. With SICEC showing significant reductions in inference time and energy cost, our work contributes to limited resource computing and sustainable AI technology. © 2020 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Edge-Cloud Synergy: Unleashing the Potential of parallel Processing for Big Data Analytics 13

Edge-Cloud Synergy: Unleashing the Potential of Parallel Pro...

引用

13th IEEE Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2022

作者： Singh, Raghubir Kiss, Tamas Department of Computer Science and Engineering Centre for Parallel Computing University of Westminster London United Kingdom

ISBN: (数字)9781665463164

ISBN: (纸本)9781665463164

If an edge-node orchestrator can partition Big Data tasks of variable computational complexity between the edge and cloud resources, major reductions in total task completion times can be achieved even at low Wide Area Network (WAN) speeds. The percentage time savings are greater with increasing task computational complexity and higher WAN speeds are required for low-complexity tasks. We demonstrate from numerical simulations that low-complexity tasks can benefit either by task partitioning between an edge node and multiple cloud servers. The orchestrator can also achieve greater time benefits by rerouting Big Data tasks directly to a single cloud resource if the balance of parameters (WAN speed and the ratio between edge and cloud processing speeds) is favourable. © 2022 IEEE.

关键词： Wide area networks

来源：评论

学校读者我要写书评

暂无评论

Optimisation and Evaluation of Breadth First Search with oneAPI/SYCL on Intel FPGAs: From Describing Algorithms to Describing Architectures 24

Optimisation and Evaluation of Breadth First Search with one...

引用

12th International Workshop on OpenCL and SYCL, IWOCL 2024

作者： Olgu, Kaan Kenter, Tobias Nunez-Yanez, Jose McIntosh-Smith, Simon School of Electrical & Electronics Engineering University of Bristol United Kingdom Paderborn Center for Parallel Computing Department of Computer Science Paderborn University Germany Dept. Electrical Engineering Linkoping University Sweden

ISBN: (纸本)9798400717901

FPGAs are often used in scientific fields to process graph algorithms due to their energy efficiency, reconfigurability, and fine-grained parallelism. However, these algorithms face challenges in memory access patterns, scalability, and programmability. The SYCL2020 implementation in the Intel oneAPI toolchain supports FPGA targets alongside SYCL2020 features like modern C++ with a single-source offloading to improve programmability. This study analysed the Breadth-First Search algorithm on Stratix 10 FPGA with the Intel oneAPI toolchain. The implementation was done in two phases. At first, we applied the typical optimisations proposed in the official guidelines alongside an automatic cache to achieve proper pipelining and improve random memory accesses performance. However, limitations occurred with fine-grained parallelism, and it was competitive only to some related work that utilised hardware-description languages or established high-level synthesis tools. For the second phase, we added bit-level representations of data in memory, banking in on-chip memory, and fine-grained control over parallel data streams. The second implementation was generally superior or on par with all compared designs, outperforming other works in 10 out of 15 tested datasets, including various synthetic RMAT and real-world datasets, with a peak performance of 1021 MTEPS. © 2024 Owner/Author.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Implementation of Basic Mathematical Operations on Openpower-ISA of Libresoc 12th

Implementation of Basic Mathematical Operations on Openpower...

引用

12th International Conference on Soft computing for Problem Solving, SocProS 2023

作者： Bilal, Maliha Simeen Chickerur, Satyadhyan Giraddi, Shantala School of Computer Science and Engineering KLE Technological University Karnataka Hubballi580031 India Centre of High-performance Computing School of Computer Science and Engineering KLE Technological University Karnataka Hubballi580031 India School of Computer Science and Engineering KLE Technological University Karnataka Hubballi580031 India

ISBN: (纸本)9789819731794

In this era, it is observed that electronic system gadgets are developed every day with the most upcoming features which are beneficial and convenient for the user. SoCs, Softcores, and Softprocessors play a vital role in developing such systems. Softcore is the digital circuitry that is implemented using logic synthesis. One such Softcore used by authors here is Libresoc. This Softcore precisely focuses on designing a SoC which contains all the functionalities of 3D GPU, CPU, and VPU. The main reason for choosing Libresoc as the base softcore is that it uses Openpower-ISA 3.0. The main objective of the proposed solution is to implement basic mathematical operations such as Addition, Subtraction, Multiplication, and Division on the decoder test cases of Openpower-ISA. This solution can act as the base for the further development of SoCs with high-end features. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： System-on-chip

来源：评论

学校读者我要写书评

暂无评论

Accelerating fluid flow in Quantum computing using GPU

Accelerating fluid flow in Quantum Computing using GPU

引用

2023 International Conference on computer, Electronics and Electrical engineering and their Applications, IC2E3 2023

作者： Dandin, Joel Chickerur, Satyadhyan Kle Technological University School of Computer Science and Engineering Hubballi India Kle Technological University Centre for High Performance Computing School of Computer Science and Engineering Hubballi India

ISBN: (纸本)9798350338003

In the last few years, quantum computing has evolved as the next big technological breakthrough. Many big companies are actively researching this area and have built open-source frameworks which can be used to develop solutions for problems that connect be solved by classical computers. One such problem is the solution of Navier-Stokes equations which takes exponential time for simulation but using quantum computers, they can be solved in polynomial time. But the current state of quantum computation does not enable everyone to test their implementation on a real quantum computer. In this situation, a simulator enables rapid prototype and testing techniques and algorithms for when the real quantum computer is ready. This paper focuses on simulating fluid flow algorithms using a quantum framework, such as Navier- Stokes equations, with the lattice Boltzmann method (LBM) used to numerically solve the relevant system of equations on both CPU and GPU to speed up the process. The quantum framework used in this paper is Qiskit, an open-source software development kit created by IBM that allows developers to create and test quantum algorithms. NVIDIA cuQuantum is an SDK of optimized libraries and tools for expediting quantum computing operations used in this paper to solve the fundamental problem of modeling large-size quantum circuits using the power of GPU © 2023 IEEE.

关键词： Navier Stokes equations

来源：评论

学校读者我要写书评

暂无评论

Learning in sinusoidal spaces with physics-informed neural networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 985-1000页

作者： Wong, Jian Cheng Ooi, Chin Chun Gupta, Abhishek Ong, Yew Soon Institute of High Performance Computing Agency for Science Technology and Research Singapore138632 Singapore School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore Singapore Institute of Manufacturing Technology Agency for Science Technology and Research Singapore138632 Singapore Agency for Science Technology and Research Singapore138632 Singapore Data Science and Artificial Intelligence Research Centre School of Computer Science and Engineering Nanyang Technological University Singapore138632 Singapore

A physics-informed neural network (PINN) uses physics-Augmented loss functions, e.g., incorporating the residual term from governing partial differential equations (PDEs), to ensure its output is consistent with fundamental physics laws. However, it turns out to be difficult to train an accurate PINN model for many problems in practice. In this article, we present a novel perspective of the merits of learning in sinusoidal spaces with PINNs. By analyzing behavior at model initialization, we first show that a PINN of increasing expressiveness induces an initial bias around flat output functions. Notably, this initial solution can be very close to satisfying many physics PDEs, i.e., falling into a localminimum of the PINN loss that onlyminimizes PDE residuals, while still being far from the true solution that jointly minimizes PDE residuals and the initial and/or boundary conditions. It is difficult for gradient descent optimization to escape from such a local minimum trap, often causing the training to stall. We then prove that the sinusoidalmapping of inputs-in an architecture we label as sf-PINN-is effective to increase input gradient variability, thus avoiding being trapped in such deceptive local minimum. The level of variability can be effectively modulated to match high-frequency patterns in the problem at hand. A key facet of this article is the comprehensive empirical study that demonstrates the efficacy of learning in sinusoidal spaces with PINNs for a wide range of forward and inversemodeling problems spanning multiple physics domains. Impact Statement-Falling under the emerging field of physicsinformed machine learning, PINN models have tremendous potential as a unifying AI framework for assimilating physics theory and measurement data. However, they remain infeasible for broad science and engineering applications due to computational cost and training challenges, especially for more complex problems. Instead of focusing on empirical demonstration of appli

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

Bidirectional LSTM-Based Energy Consumption Forecasting:Advancing AI-Driven Cloud Integration for Cognitive City Energy Management

引用

computers, Materials & Continua 2025年第5期83卷 2907-2926页

作者： Sheik Mohideen Shah Meganathan Selvamani Mahesh Thyluru Ramakrishna Surbhi Bhatia Khan Shakila Basheer Wajdan Al Malwi Mohammad Tabrez Quasim Department of Computer Science and Engineering Srinivasa Ramanujan CentreSASTRA Deemed UniversityKumbakonam612001India Department of Computer Science and Engineering Faculty of Engineering and TechnologyJAIN(Deemed-to-be University)Bangalore562112India School of Science Engineering and EnvironmentUniversity of SalfordManchesterM54WTUK University Centre for Research and Development Chandigarh UniversityMohali140413PunjabIndia Centre for Research Impact and Outcome Chitkara University Institute of Engineering and TechnologyChitkara UniversityRajpura140401PunjabIndia Department of Information Systems College ofComputer and Information SciencePrincessNourah bintAbdulrahmanUniversityP.O.Box 84428Riyadh11671Saudi Arabia College of Computer Science Informatics and Computer Systems Department King Khalid UniversityAbha61421Saudi Arabia Department of Computer Science and Artificial Intelligence College of Computing and Information TechnologyUniversity of BishaBisha61922Saudi Arabia

Efficient energy management is a cornerstone of advancing cognitive cities,where AI,IoT,and cloud computing seamlessly integrate to meet escalating global energy *** this context,the ability to forecast electricity consumption with precision is vital,particularly in residential settings where usage patterns are highly variable and *** study presents an innovative approach to energy consumption forecasting using a bidirectional Long Short-Term Memory(LSTM)*** a dataset containing over twomillionmultivariate,time-series observations collected froma single household over nearly four years,ourmodel addresses the limitations of traditional time-series forecasting methods,which often struggle with temporal dependencies and non-linear *** bidirectional LSTM architecture processes data in both forward and backward directions,capturing past and future contexts at each time step,whereas existing unidirectional LSTMs consider only a single temporal *** design,combined with dropout regularization,leads to a 20.6%reduction in RMSE and an 18.8%improvement in MAE over conventional unidirectional LSTMs,demonstrating a substantial enhancement in prediction accuracy and *** to existing models—including SVM,Random Forest,MLP,ANN,and CNN—the proposed model achieves the lowest MAE of 0.0831 and RMSE of 0.2213 during testing,significantly outperforming these *** results highlight the model’s superior ability to navigate the complexities of energy usage patterns,reinforcing its potential application in AI-driven IoT and cloud-enabled energy management systems for cognitive *** integrating advanced machine learning techniqueswith IoT and cloud infrastructure,this research contributes to the development of intelligent,sustainable urban environments.

关键词： Deep learning bidirectional LSTM energy consumption forecasting time-series analysis predictive modeling machine learning in energy management

来源：评论

学校读者我要写书评

暂无评论

BERD+: A Generic Sequential Recommendation Framework by Eliminating Unreliable Data with Item- and Attribute-level Signals

引用

ACM Transactions on Information Systems 2023年第2期42卷 1-33页

作者： Sun, Yatong Yang, Xiaochun Sun, Zhu Wang, Bin School of Computer Science and Engineering Northeastern University School of Computing Macquarie University China Australia Institute of High Performance Computing Centre for Frontier AI Research A*STAR Singapore School of Computer Science and Engineering Northeastern University China National Frontiers Science Center for Industrial Intelligence and Systems Optimization China Key Laboratory of Data Analytics and Optimization for Smart Industry [Northeastern University Ministry of Education China

A preliminary work of our study was published at IJCAI’21 [47], which is substantially extended in the following aspects: (1) In Section 1, we analyze the necessity of introducing item attributes for detecting unreliable instances, together with the problems and challenges that attributes may bring in. (2) In Section 2, we add discussions about the limitations of existing attribute-aware recommender systems (Section 2.2) and denoising methods (Section 2.3) in the context of detecting unreliable instances. (3) In Section 4.2, we further conduct an in-depth analysis at the attribute level to demonstrate the capability of attributes for rectifying instance loss and uncertainty, as well as the disturbance caused by attributes. (4) We generalize BERD to a generic framework BERD+ in Section 5.1, equipped with novel modules, i.e., HU-GCN (Section 5.2) and EPE (Section 5.4), which properly incorporate item attributes while reducing their disturbance for rectifying instance uncer tainty (Section 5.5) and loss (Section 5.6). The generic BERD+ can be flexibly plugged into existing SRSs for performance enhanced recommendation via eliminating unreliable data. (5) In Section 6.2, we apply our BERD+ framework to seven state-of-the-art SRSs on five real-world datasets to illustrate its superiority. (6) To avoid unfair comparison caused by item attributes, we build and compare with the baseline that combines the original BERD and an advanced attribute-aware recommender system, KSR [19]. (7) For more comprehensive comparison, in Section 6.2.2, we compare BRED+ with two state-of-the-art denoising approaches;in Section 6.2.3, to examine the efficacy of HU-GCN and EPE, we compare HU-GCN with various attribute embedding techniques, i.e., variants of graph neural networks, and compare EPE with different attribute fusing methods, i.e., adding, concatenation, and weighted sum. (8) In Section 6.2.4, a detailed ablation study is conducted to verify the effectiveness of each module of BERD+. (

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Few-shot Prompting for Pairwise Ranking: An Effective Non-Parametric Retrieval Model

Few-shot Prompting for Pairwise Ranking: An Effective Non-Pa...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Sinhababu, Nilanjan Parry, Andrew Ganguly, Debasis Samanta, Debasis Mitra, Pabitra Centre for Computational and Data Sciences IIT Kharagpur India School of Computing Science University of Glasgow United Kingdom Department of Computer Science and Engineering IIT Kharagpur India

ISBN: (纸本)9798891761681

A supervised ranking model, despite its effectiveness over traditional approaches, usually involves complex processing - typically multiple stages of task-specific pre-training and fine-tuning. This has motivated researchers to explore simpler pipelines leveraging large language models (LLMs) that can work in a zero-shot manner. However, since zero-shot inference does not make use of a training set of pairs of queries and their relevant documents, its performance is mostly worse than that of supervised models, which are trained on such example pairs. Motivated by the existing findings that training examples generally improve zero-shot performance, in our work, we explore if this also applies to ranking models. More specifically, given a query and a pair of documents, the preference prediction task is improved by augmenting examples of preferences for similar queries from a training set. Our proposed pairwise few-shot ranker demonstrates consistent improvements over the zero-shot baseline on both in-domain (TREC DL) and out-domain (BEIR subset) retrieval benchmarks. Our method also achieves a close performance to that of a supervised model without requiring any complex training pipeline. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

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：