检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Sah, Suwesh Prasad Chinara, Suchismita Dept. Computer Science and Engineering National Institute of Technology Rourkela India

Autonomous driving in high-speed racing, as opposed to urban environments, presents significant challenges in scene understanding due to rapid changes in the track environment. Traditional sequential network approaches may struggle to meet the real-time knowledge and decision-making demands of an autonomous agent covering large displacements in a short time. This paper proposes a novel baseline architecture for developing sophisticated models capable of true hardware-enabled parallelism, achieving neural processing speeds that mirror the agent’s high velocity. The proposed model (Parallel Perception Network (PPN)) consists of two independent neural networks, segmentation and reconstruction networks, running parallelly on separate accelerated hardware. The model takes raw 3D point cloud data from the LiDAR sensor as input and converts it into a 2D Bird’s Eye View Map on both devices. Each network independently extracts its input features along space and time dimensions and produces outputs parallelly. The proposed method’s model is trained on a system with two NVIDIA T4 GPUs, using a combination of loss functions, including edge preservation, and demonstrates a 2x speedup in model inference time compared to a sequential configuration. Implementation is available at: github/ParallelPerceptionNetwork. Learned parameters of the trained networks are provided at: huggingface/ParallelPerceptionNetwork. © 2024, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A 4x4 MIMO Antenna Array for the Wi-Fi7Access Point on a Metal Wall Application

A 4x4 MIMO Antenna Array for the Wi-Fi7Access Point on a Met...

引用

International Symposium on Antennas and Propagation (ISAP)

作者： Hsin-Lung Su You-Sheng Zhan Dept. of Computer and Communication National Pingtung University Taiwan Dept. of Computer Science and Information Engineering National Pingtung University Taiwan

ISBN: (数字)9798350364774

ISBN: (纸本)9798350364781

4×4 CPW MIMO Antenna array with an artificial magnetic conductor (AMC) reflector is presented for Wi-Fi7 access point on a metal wall application in the study. The 10-dB impedance of the integrated MIMO antenna array covers the Wi-Fi 7 operating bands, and the transmission coefficients are under -15 dB in the whole Wi-Fi7 bands. This proposed integrated MIMO antenna is low profile and has good total radiation efficiency (over 60%). The height of the integrated MIMO antenna array is 6.325 mm.

关键词： Transmitting antennas Metals Directive antennas Reflector antennas Conductors Coplanar waveguides Impedance Wireless fidelity Antenna arrays Antenna radiation patterns

来源：评论

学校读者我要写书评

暂无评论

Know Your Enemy To Save Cloud Energy: Energy-Performance Characterization of Machine Learning Serving

Know Your Enemy To Save Cloud Energy: Energy-Performance Cha...

引用

IEEE Symposium on High-Performance computer Architecture

作者： Junyeol Yu Jongseok Kim Euiseong Seo Dept. of Computer Science and Engineering Sungkyunkwan University

The proportion of machine learning (ML) inference in modern cloud workloads is rapidly increasing, and graphic processing units (GPUs) are the most preferred computational accelerators for it. The massively parallel computing capability of GPUs is well-suited to the inference workloads but consumes more power than conventional CPUs. Therefore, GPU servers contribute significantly to the total power consumption of a data center. However, despite their heavy power consumption, GPU power management in cloud-scale has not yet been actively researched. In this paper, we reveal three findings about energy efficiency of ML inference clusters in the cloud. ❶ GPUs of different architectures have comparative advantages in energy efficiency to each other for a set of ML models. ❷ The energy efficiency of a GPU set may significantly vary depending on the number of active GPUs and their clock frequencies even when producing the same level of throughput. ❸ The service level objective(SLO)-blind dynamic voltage and frequency scaling (DVFS) driver of commercial GPUs maintain an immoderately high clock frequency. Based on these implications, we propose a hierarchical GPU resource management approach for cloud-scale inference services. The proposed approach consists of energy-aware cluster allocation, intra-cluster node scaling, intra-node GPU scaling and GPU clock scaling schemes considering the inference service architecture hierarchy. We evaluated our approach with its prototype implementation and cloud-scale simulation. The evaluation with real-world traces showed that the proposed schemes can save up to 28.3% of the cloud-scale energy consumption when serving five ML models with 105 servers having three different kinds of GPUs.

关键词： Energy consumption Power demand Graphics processing units Prototypes computer architecture Machine learning Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Gotcha: Real-Time Video Deepfake Detection via Challenge-Response

Gotcha: Real-Time Video Deepfake Detection via Challenge-Res...

引用

IEEE European Symposium on Security and Privacy (EuroS&P)

作者： Govind Mittal Chinmay Hegde Nasir Memon Dept. of Computer Science and Engineering Tandon School of Engineering New York University USA

ISBN: (数字)9798350354256

ISBN: (纸本)9798350354263

With the rise of AI-enabled Real-Time Deepfakes (RTDFs), the integrity of online video interactions has become a growing concern. RTDFs have now made it feasible to replace an imposter's face with their victim in live video interactions. Such advancement in deepfakes also coaxes detection to rise to the same standard. However, existing deepfake detection techniques are asynchronous and hence ill-suited for RTDFs. To bridge this gap, we propose a challenge-response approach that establishes authenticity in live settings. We focus on talking-head style video interaction and present a taxonomy of challenges that specifically target inherent limitations of RTDF generation pipelines. We evaluate representative examples from the taxonomy by collecting a unique dataset comprising eight challenges, which consistently and visibly degrades the quality of state-of-the-art deepfake generators. These results are corroborated both by humans and a new automated scoring function, leading to 88.6% and 80.1% AUC, respectively. The findings under-score the promising potential of challenge-response systems for explainable and scalable real-time deepfake detection in practical scenarios. We provide access to data and code at https://***/mittalgovind/GOTCHA-Deepfakes.

关键词： Deepfakes Codes Taxonomy Pipelines Streaming media Real-time systems Robustness

来源：评论

学校读者我要写书评

暂无评论

Modular Soft-switched PV Converter with an Efficiency Optimization Scheme for High Frequency Linking Power Balancers

Modular Soft-switched PV Converter with an Efficiency Optimi...

引用

Annual IEEE Conference on Applied Power Electronics Conference and Exposition (APEC)

作者： Kajanan Kanathipan John Lam Dept. of Electrical Engineering and Computer Science York University Toronto Canada

A modular photovoltaic (PV) step-up converter with embedded high frequency power balancers that utilize an interlinking controller capable of power efficiency optimization over a wide operating range for Medium Voltage (MV) DC distribution is proposed in this paper. The converter utilizes an integrated boost-CLL resonant circuit with power balancing active voltage quadruplers (AVQ) to achieve step-up voltage conversion, individual maximum power point tracking (MPPT) and modular power balancing. The proposed interlinking power balancer controller optimizes the transmission of mismatched power while simultaneously achieving soft-switching operation on all AVQ circuits. The modular nature of the system allows for additional modules to be added without requiring parameter changes to the existing system. Analysis of the proposed converter and power balancing technique along with 10kW, 10kV-output system simulation and preliminary results on a scaled down 500W, 850V-output system are provided to highlight the features of the proposed system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Efficient Deep Learning Approach for Jute Pest Classification Using Transfer Learning

An Efficient Deep Learning Approach for Jute Pest Classifica...

引用

International Conference on Electrical engineering and Information & Communication Technology (ICEEICT)

作者： Muhammad Tanvirul Islam Md. Sadekur Rahman Dept. of Computer Science and Engineering Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798350385779

ISBN: (纸本)9798350385786

Jute is considered as one of the most vital crops in the world. For some countries jute is the principal source of earnings and GDP. One of the primary elements influencing jute yield is jute pests. Accurate pest identification makes it possible to take prompt preventative action to minimize financial losses. Considering the fact, to classify jute pests, the study suggests different jute pest classification models, which are based on transfer learning. The best model offers high performance and resilience. A VCI-validated dataset comprising 7235 images has been utilized in the analysis. The dataset encompasses images classified into 17 distinct jute pest classes. The dataset is already divided into three categories train, test and validation. To increase the dataset size, data augmentation is applied to the training set. To improve performance, all the models were integrated with the transfer learning model. VGG 16, ResNetl0l, DenseNet201, InceptionV3, Xception, and MobileN etV2 were used to train the parameters on the publicly available ImageN et dataset followed by some customized dense layers. The models were assessed using different types of metrics, including confusion matrix, F1 score, precision, and recall. Compared to other models DenseNet201 outclassed other models, acquiring 97% accuracy. The fundamental information and technical support for jute pest classification are provided by this study.

关键词： Training Measurement Deep learning Economic indicators Transfer learning Crops Data augmentation

来源：评论

学校读者我要写书评

暂无评论

Developing a Self-Explanatory Transformer

Developing a Self-Explanatory Transformer

引用

IEEE/ACM Symposium on Edge Computing (SEC)

作者： Rasha Karakchi Ryan Karbowniczak Dept. of Computer Science and Engineering University of South Carolina Columbia USA

ISBN: (数字)9798350378283

ISBN: (纸本)9798350378290

While IoT devices provide significant benefits, their rapid growth results in larger data volumes, increased complexity, and higher security risks. To manage these issues, techniques like encryption, compression, and mapping are used to process data efficiently and securely. General-purpose and AI platforms handle these tasks well, but mapping in natural language processing is often slowed by training times. This work explores a self-explanatory, training-free mapping transformer based on non-deterministic finite automata, designed for Field-Programmable Gate Arrays (FPGAs). Besides highlighting the advantages of this proposed approach in providing real-time, cost-effective processing and dataset-loading, we also address the challenges and considerations for enhancing the design in future iterations.

关键词： Training Automata Transformers Real-time systems Natural language processing Encryption Security Internet of Things Field programmable gate arrays Edge computing

来源：评论

学校读者我要写书评

暂无评论

Cardiovascular Disease Prediction Utilizing Machine Learning and Feature Selection with Clonal Selection Algorithm

Cardiovascular Disease Prediction Utilizing Machine Learning...

引用

Sustainable Technologies for Industry 4.0 (STI), International Conference on

作者： Wahidur Rahman Riasaad Haque Aneek Md. Moinuddin Md. Shadman Sakib Md. Sadiq Iqbal Mohammad Motiur Rahman Dept. of Computer Science and Engineering Uttara University Uttara Dhaka Bangladesh Dept. of Computer Science and Engineering Z.H. Sikder University of Science and Technology Bhedarganj Shariatpur Dept. of Computer Science and Engineering University of Information Technology and Sciences Dhaka Bangladesh Dept. of Computer Science and Engineering Bangladesh University Adabor Bangladesh Dept. of Computer Science and Engineering Mawlana Bhashani Science and Technology University Tangail Bangladesh

A substantial and rising number of patients suffer from cardiovascular diseases, including heart attacks, heart failure, and other related illnesses. This case surge places increasing pressure on healthcare professionals and administrators while patients grapple with growing medical costs. To address these challenges, an automated system is necessary within the healthcare sector. In this paper, we present a cardiovascular health monitoring system that incorporates Machine Learning techniques. The study employed feature selection techniques on a secondary dataset. The study has used a nature-inspired technique, namely the Clonal Selection Algorithm (CSA) and Maximum Relevance Minimum Redundancy (mRMR) technique, to identify the most prominent feature in the detection of cardio-vascular illness. This approach was combined with a collection of Machine Learning classifiers. A group of experimental data has been listed to assess the suggested model's effectiveness. The research findings indicated a maximum accuracy rate of 100% when employing the proposed algorithms and orientations. The study also discusses the performance analysis for CSA and mRMR using a set of performance evaluation matrices. Based on the obtained results, it can be inferred that the suggested model will likely exhibit a high level of effectiveness in identifying cardiovascular diseases.

关键词：

来源：评论

学校读者我要写书评

暂无评论

COMPOSITIONAL GENERATIVE INVERSE DESIGN

arXiv

引用

arXiv 2024年

作者： Wu, Tailin Maruyama, Takashi Wei, Long Zhang, Tao Du, Yilun Iaccarino, Gianluca Leskovec, Jure Dept. of Engineering Westlake University China NEC Laboratories Europe Germany Dept. of Computer Science MIT United States Dept. of Mechanical Engineering Stanford University United States Dept. of Computer Science Stanford University United States

Inverse design, where we seek to design input variables in order to optimize an underlying objective function, is an important problem that arises across fields such as mechanical engineering to aerospace engineering. Inverse design is typically formulated as an optimization problem, with recent works leveraging optimization across learned dynamics models. However, as models are optimized they tend to fall into adversarial modes, preventing effective sampling. We illustrate that by instead optimizing over the learned energy function captured by the diffusion model, we can avoid such adversarial examples and significantly improve design performance. We further illustrate how such a design system is compositional, enabling us to combine multiple different diffusion models representing subcomponents of our desired system to design systems with every specified component. In an N-body interaction task and a challenging 2D multi-airfoil design task, we demonstrate that by composing the learned diffusion model at test time, our method allows us to design initial states and boundary shapes that are more complex than those in the training data. Our method generalizes to more objects for N-body dataset and discovers formation flying to minimize drag in the multi-airfoil design task. Project website and code can be found at https://***/AI4science-WestlakeU/cindm. © 2024, CC BY.

关键词： Airfoils

来源：评论

学校读者我要写书评

暂无评论

“Precision Agriculture Farming” Enhancing Farming Efficiency through Technology Integration

“Precision Agriculture Farming” Enhancing Farming Efficien...

引用

Advances in Computing, Communication and Networking (ICAC2N), International Conference on

作者： Prashant Sharma Pratyaksh Kumar Praveen Kumar Singh Arun Kumar Rana Dept. Of Computer Science And Engineering Galgotias College of Engineering And Technology Greater Noida India

ISBN: (数字)9798350356816

ISBN: (纸本)9798350356823

Precision agriculture, or precision farming, or smart farming, refers to a contemporary approach that combines technology and data analytics with agricultural practices. This research paper examines the concept of precision agriculture, its principles, advantages, disadvantages, and potential to transform farming practices. Through the use of various technologies including GPS, sensors, drones, and data analytics, precision agriculture seeks to maximize resource efficiency, enhance crop yields, reduce the environmental footprint, and boost overall profitability for the farm. The paper also addresses case studies and real-life implementations of precision agriculture and illustrates its effectiveness in various sectors of agriculture. It further identifies the challenges and constraints of adopting precision agriculture and proposes possible solutions to overcome these challenges. This paper generally offers insight into the revolutionary potential of precision agriculture in reshaping the future of farming.

关键词： Precision agriculture Training Data analysis Profitability Transforms Internet of Things Artificial intelligence Farming Global Positioning system Drones

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：