检索结果-内蒙古大学图书馆

10th IEEE Smart World Congress, SWC 2024

作者： You, Feiran Du, Hongyang Kang, Jiawen Ni, Wei Niyato, Dusit Jamalipour, Abbas The University of Sydney School of Electrical and Computer Engineering SydneyNSW2006 Australia SydneyNSW2122 Australia University of Hong Kong Department of Electrical and Electronic Engineering Pok Fu Lam Hong Kong Guangdong University of Technology School of Automation China Nanyang Technological University College of Computing and Data Science Singapore

ISBN: (纸本)9798331520861

The rapid growth of Artificial Intelligence-Generated Content (AIGC) services has led to increased mobile user participation in related computations and interactions. This development has enabled AI-generated characters to interact with Virtual Reality (VR) users in real time, making the VR experience more interactive and personalized. In this paper, we consider an MEC system where VR users engage in AIGC services, focusing on the Generative Diffusion Model (GDM)based image generation tasks. Specifically, VR users initiate requests for computing resources, while computation offloading distributes the processing load across the MEC system. To manage AIGC edge computation offloading and cloudlet-VR user connections jointly, a Data Center Operator (DCO) employs a centralized Proximal Policy Optimization (PPO) algorithm. To protect VR users' privacy while preserving PPO functionality, we employ the Generative Diffusion Model (GDM), specifically the Denoising Diffusion Implicit Model (DDIM), which first introduces noise to the PPO state, then conducts a denoising process to recover the state information. We further employ Inverse Reinforcement Learning (IRL) to infer rewards for the recovered states, using expert demonstrations trained by the PPO. The similarity between PPO-generated rewards and IRL-inferred rewards is then computed. Simulation results demonstrate that our proposed approach successfully achieves computation offloading while protecting VR users' privacy within the PPO centralized management framework. © 2024 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Advancements in Cricket Shot Detection: Integrating Human Pose Estimation and Deep Learning for Automated Analysis

Advancements in Cricket Shot Detection: Integrating Human Po...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Ks Sreekar Datta G Narasimha Naidu Rahul Dasari T V Jayakumar Dept. Of Computer Science and Engineering Amrita School Of Computing Amrita Vishwa Vidyapeetham Amritapuri India Dept. Of Computer Science and Applications Amrita School Of Computing Amrita Vishwa Vidyapeetham Amritapuri India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Cricket, a sport loved worldwide, often presents challenges for aspiring players due to limited access to infrastructure and coaching resources. Cricket’s automated learning solutions are still largely unexplored, despite its popularity. By concentrating on the application of cricket shot detection, a crucial component of the game, this research fills this gap. One essential component of computer vision that becomes apparent in deciphering the subtleties of cricket shots is human stance estimate. Pose estimate provides a detailed examination of batting skills and footwork by tracking important areas on the batsman’s body, such as joints and limbs. MoveNet is a significant framework in this domain that has garnered accolades for its real-time performance and lightweight architecture, which enable accurate identification of human poses, which is a crucial requirement in the dynamic world of cricket. This study explores the combination of deep learning techniques and human pose estimate for cricket shot recognition. More precise and reliable cricket shot identification and classification are made possible by the model’s ability to extract complex patterns and features from the input data by utilizing deep learning techniques. Our goal is to automatically identify and classify cricket shots based on players’ minor motions through theoretical investigation, empirical validation, and real-world applications. Through the integration of cutting-edge technologies, such as deep learning, with traditional cricket analysis, our goal is to gain a deeper understanding of the game and discover new avenues for sports analytics, coaching, and broadcasting.

关键词： Deep learning Training Analytical models Accuracy Refining Pose estimation Games Transforms Feature extraction Sports

来源：评论

学校读者我要写书评

暂无评论

A Code Completion Approach Combining Pointer Network and Transformer-XL Network 19th

A Code Completion Approach Combining Pointer Network and Tr...

引用

19th EAI International Conference on Collaborative computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Zhang, Xiangping Liu, Jianxun Long, Teng Hu, Haize School of Computer Science and Engineering Hunan University of Science and Technology Hunan Xiangtan411100 China Hunan Key Lab. for Services Computing and Novel Software Technology Hunan University of Science and Technology Hunan Xiangtan411100 China

ISBN: (纸本)9783031545207

Code completion is an integral component of modern integrated development environments, as it not only facilitates the software development process but also enhances the quality of software products. By leveraging large-scale codes to learn the probability distribution among code token units, deep learning methods have demonstrated significant improvements in the accuracy of token unit recommendations. However, the effectiveness of code completion with deep learning techniques is hindered by information loss. To alleviate the above problem, we proposed a code language model which combines the pointer network and Transformer-XL network to overcome the limitations of existing approaches in code completion. The proposed model takes as input the original code fragment and its corresponding abstract syntax tree and leverages the Transformer-XL model as the basis model for capturing long-term dependencies. Furthermore, we integrate a pointer network as a local component to predict the out-of-vocabulary words. The proposed method is evaluated on real PY150 and JS150 datasets. The comparative experimental results demonstrate the effectiveness of our model in improving the accuracy of the code completion task at the token unit level. © ICST Institute for computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

Automatic Topic Identification in Destination Reviews Using Hybrid Topic Modelling Techniques

Automatic Topic Identification in Destination Reviews Using ...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Majji Nitin Sai Neelam Leeladhar Royal Narayana Chakravarthi Nagavarapu Siva Sai Deepa Gupta Manju Venugopalan Dept. of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Nowadays, with user-generated material flooding digital platforms, it is crucial to analyze it thoroughly because it is a huge data source that provides insightful information for both research and economic objectives. Making use of the unstructured destination reviews that are a significant part of this data pool found on the social media platforms, allows for great insight into the preferences of tourists and makes it easier to improve the tourism experience. Using a blend of LDA, LLM, and clustering methods, the proposed model efficiently extracts topics from the vast number of unstructured reviews. For instance the model’s efficiency as it undergoes performance review using coherence and silhouette scores, has shown good results, especially in the case of Coorg, a coherence score of 0.48 for LDA. However, the superb performance was revealed in GPT2 by the silhouette score of 0.57 for Coorg. It reflects that the model is quite capable of extracting intrinsic information from complex texts that refer to Coorg. With the use of this method, significant information can be extracted from even the most intricate and contextually rich evaluations, offering players in the tourist industry deep insights, and streamlining their decision-making processes. In the end, this gives stakeholders the ability to efficiently match offerings with visitor preferences.

关键词： Industries Reviews Social networking (online) Computational modeling Soft sensors Coherence Data models Data mining Stakeholders Floods

来源：评论

学校读者我要写书评

暂无评论

An In-Depth Exploration of Image Captioning Training Approaches and Performance Analysis

An In-Depth Exploration of Image Captioning Training Approac...

引用

International Conference for Convergence of Technology (I2CT)

作者： K Kushal Madhav Manoj Kevansh Reddy Priyanka C Nair Dept. of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798350394474

ISBN: (纸本)9798350394481

In the rapidly evolving landscape of computer vision, image captioning has emerged as a challenging task. This report explores the techniques involved in image captioning using deep learning techniques, especially the encoder-decoder framework. Models such as CNN-LSTM, CNN-GRU,Xception – YOLO v4, GIT Based Model are used. Recognition is given to the significance of ample and well-annotated datasets in teaching the algorithms to understand the complex relationships between visual elements and textual descriptions. Along with traditional evaluation metrics like BLEU score, this study also employs metrics such as METEOR, ROUGE-L, and SPICE to compare performance between models. The findings highlight on the impact of deep learning in enabling computers to generate captions for diverse visual content.

关键词： Measurement Deep learning YOLO Training Visualization Computational modeling SPICE

来源：评论

学校读者我要写书评

暂无评论

Deep reinforcement learning with time-scale invariant memory

arXiv

引用

arXiv 2024年

作者： Kabir, Md Rysul Mochizuki-Freeman, James Tiganj, Zoran Department of Computer Science Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States

The ability to estimate temporal relationships is critical for both animals and artificial agents. Cognitive science and neuroscience provide remarkable insights into behavioral and neural aspects of temporal credit assignment. In particular, scale invariance of learning dynamics, observed in behavior and supported by neural data, is one of the key principles that governs animal perception: proportional rescaling of temporal relationships does not alter the overall learning efficiency. Here we integrate a computational neuroscience model of scale invariant memory into deep reinforcement learning (RL) agents. We first provide a theoretical analysis and then demonstrate through experiments that such agents can learn robustly across a wide range of temporal scales, unlike agents built with commonly used recurrent memory architectures such as LSTM. This result illustrates that incorporating computational principles from neuroscience and cognitive science into deep neural networks can enhance adaptability to complex temporal dynamics, mirroring some of the core properties of human learning. © 2024, CC BY.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

License Plate Detection using YOLO v8 and Performance Evaluation of EasyOCR, PaddleOCR and Tesseract

License Plate Detection using YOLO v8 and Performance Evalua...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： P Praneeth Reddy P Sai Shruthi P Himanshu Tripty Singh Dept. of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

License plate recognition is critical to the functioning of the control and surveillance systems. Vehicle identification is based on number plates, which are made up of a certain arrangement of letters and digits. However, manually identifying every number plate is a laborious and difficult task. The majority of earlier automatic number plate recognition (ANPR) systems were restricted in how they could operate. These conditions included being limited to fixed illumination, background noise in the vehicle image. The primary goal of this project is to develop a reliable number plate recognition model that functions with the help of YOLO v8 for license plate segmentation followed with a comparison model of EasyOCR, PaddleOCR, Tesseract models based on best model for character recognition of license plate. The steps involved include processing of vehicle image, segmentation of license plate from the image, followed by internal segmentation of characters on the plate and pattern matching. Data pre-processing is finally performed in order to read and process the license plate. Problems like background noise in image like characters apart from the license plate and number-based characters being wrongly predicted as alphabets have been tackled.

关键词： YOLO Performance evaluation Image segmentation Surveillance Lighting Control systems Reliability Background noise License plate recognition Pattern matching

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Deep Learning Methods for Lesion Segmentation in Gastrointestinal Endoscopic Images

Comparative Analysis of Deep Learning Methods for Lesion Seg...

引用

Women in computing (InCoWoCo), International Conference for

作者： K Sai Chandana Reddy K Dheeraj Reddy L E Sree Sai Praneeth Goud Tripty Singh K Afnaan Dept. of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798331518943

ISBN: (纸本)9798331518950

Gastrointestinal (GI) disorders present diagnostic obstacles, necessitating accurate imaging procedures. Conventional techniques for evaluating endoscopic images are mostly laborious and arbitrary. An innovative method is proposed to improve GI endoscopic image evaluation using three cuttingedge deep learning models, including Seg Net, U-Net, and Deep Lab, for lesion segmentation. Exploiting the recently published Kvasir-SEG dataset, which comprises segmentation masks and accompanied gastrointestinal polyp images, this work attempts to further the field of polyp analysis and detection research. The study aims to enhance diagnostic precision and optimize the analytical process by utilizing deep learning approaches and datasets that are annotated. It will eventually contribute for enhanced patient outcomes as well as healthcare productivity.

关键词： Deep learning Image segmentation Analytical models Accuracy Computational modeling Predictive models Gastrointestinal tract Software Lesions Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Improvising energy efficiency of heterogeneous servers using AWS services and Machine Learning approaches

Improvising energy efficiency of heterogeneous servers using...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： P Praneeth Reddy P Sai Shruthi P Tharneesh B.M Beena Dept. of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

The rapid expansion of data centers has led to a substantial increase in energy consumption, prompting a need for more energy-efficient solutions. This paper presents a comprehensive approach to enhancing the energy efficiency of heterogeneous servers by leveraging Amazon Web Services (AWS) and Machine Learning (ML) techniques. Our proposed solution involves developing an ML model that uses a dataset of VM cloud which consists of various parameters like the cloud ID, timestamp, CPU performance, memory usage, network traffic, energy efficiency and so on. The model uses regression algorithms like Linear Regression, Decision Tree Regressor, K-Neighbor Regressor, Polynomial Regression, and AdaBoost Regressor to predict the energy-efficiency of a system based on the relevant features. The model will be evaluated on the basis of Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, R2 Score, and Mean Absolute Percentage Error and Linear Regression model is noted to provide the least error scores. This model helps in reducing server workloads and dynamically manages resource allocation to minimize energy usage while maintaining optimal performance. The developed ML model is then deployed on the AWS EC2 free tier instance by encapsulating the ML models within a Docker container. This study provides a scalable and economical approach to enhance the energy efficiency in the data centers, with a potential impact on the wide range of cloud -based infrastructures.

关键词： Data centers Computational modeling Linear regression Machine learning Containers Energy efficiency Servers Resource management Regression tree analysis Context modeling

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approaches for Lung Cancer Detection: A Transfer Learning Perspective

SSRN

引用

SSRN 2024年

作者： Pandiyarajan, Pandiselvam Niteesh, V. Sriram, S. Durga Sai Amrutha, T. Maheswaran, Baskaran Department of Computer Science and Engineering School of Computing Kalasalingam Academy of Research and Education Tamilnadu Krishnankoil India

Lung cancer aids a substantial global health issue, requiring the enlargement of precise and efficient diagnostic techniques. For the purpose of classification of different forms of lung cancer using CT scans, this study investigated the deep learning techniques' potential, specifically transfer learning. In order to produce reliable classifiers, we performed feature extraction and fine-tuning using models that have already been trained like VGG16, MobileNet, DenseNet, Inception V3, and AlexNet. According on our findings, the suggested strategy, performed exceptionally well, achieving an accuracy rate of 99.09%. Moreover, we investigated the impact of different activation functions on model performance in medical image analysis tasks. Despite the limitations associated with the availability of training data, our study highlights the significance of the diagnosis and treatment of lung cancer. Future research should focus on obtaining larger datasets and integrating comprehensive clinical information to further enhance the performances of deep-learning-based approaches in lung cancer treatment. © 2024, The Authors. All rights reserved.

关键词： computerized tomography

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：