检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第5期5卷 2331-2341页

作者： Yu, Qianzi Zhu, Kai Wang, Wen Cao, Yang Kang, Yu University of Science and Technology of China Department of Automation School of Information Science and Technology Hefei230026 China Hefei Comprehensive National Science Center Institute of Artificial Intelligence Hefei230039 China Zhejiang University College of Computer Science and Technology State Key Lab of CAD&CG Hangzhou310027 China University of Science and Technology of China School of Information Science and Technology Hefei230026 China

Incremental object detection (IOD) aims to achieve simultaneous prediction of old and new samples on localization and classification when new concepts are provided. It is a challenging task due to the need for a joint interpretation of semantic and spatial information. While the existing work accomplishes the detector generalization with the help of annotated sample classes and bounding boxes, this article presents one astonishing finding on the unnecessity of new bounding boxes, which will significantly reduce the annotation cost and condition constraints in IOD. To enable the incremental detection process with image-level labels, we propose a multibranch decoupling scheme in which the representation, classification, and regression branches are customized to accommodate new concepts with class labels with different semantic levels. The regression branch is first frozen to ensure generalizable localization while maintaining the stability of representation optimization. Then, the representation branch is gradient-inverted to estimate the distribution drift from the old object-level set to the new image-level one, improving the effectiveness of feature updates in the absence of homogeneous semantic and spatial supervision. Finally, the classification branch is calibrated to balance the contribution of different images to the gradient, in which the frequency ratios of old instances and new images are calculated as a postprocessing factor for the logits margin. Extensive experiments on two standard incremental detection benchmarks demonstrate the above-par recognition and localization performance while reducing the burden on new class annotation, outperforming state-of-the-art methods by 2% and 3%, respectively. © 2023 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

A semantic driven model for extraction of text using TF-IDF, SFLA and XGBoost

引用

International Journal of Information technology (Singapore) 2025年 1-6页

作者： Madhumala, R.B. Vineetha, B. Shree, M. Rajani Sanjesh, Riya Charanraj, B.R. Department of Computer Science and Engineering Dayanand Sagar Academy of Technology and Management (DSATM) Bengaluru India Computer Science and Engineering Presidency University Bangalore India Department of CSE BNM Institute of Technology Bengaluru India Computer Science PES University BANGALORE URBAN India

As people increasingly express opinions, offer feedback, and share suggestions on websites, e-forums, and blogs, consumers have come to rely heavily on online product reviews before making purchases or using services. Some spammers deliberately manipulate reviews to either enhance or discredit products, leading consumers to make misguided decisions based on these misleading reviews. To address this issue, a hybrid approach combining TF-IDF (term frequency-inverse document frequency) with the shuffled frog leaping algorithm (SFLA) is proposed. This approach aims to reduce the high dimensionality of feature sets and select optimized subsets of features. Finally, the sentiments from online reviews are classified into genuine and misleading by using Extreme Gradient Boosting (XGBoost) classifier. By enhancing review credibility, the proposed approach helps customers identify authentic feedback and improves classification accuracy. The performance of the proposed approach is assessed and validated with existing studies based on accuracy (%), precision (%) and recall (%). The results indicate that this hybrid feature selection approach achieves an optimized subset of features, leading to superior classification accuracy. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： Classification SFLA Text extraction TF-IDF XGBoost

来源：评论

学校读者我要写书评

暂无评论

A Transformer Based on Feedback Attention Mechanism for Diagnosis of Coronary Heart Disease Using Echocardiographic Images

引用

computers, Materials & Continua 2025年第5期83卷 3435-3450页

作者： Chunlai Du Xin Gu Yanhui Guo Siqi Guo Ziwei Pang Yi Du Guoqing Du School of Information Science and Technology North China University of TechnologyBeijing100144China Department of Computer Science University of Illinois SpringfieldSpringfieldIL 62703USA Department of Ultrasound Sun Yat-sen Memorial HospitalSun Yat-sen UniversityGuangzhou510120China

Coronary artery disease is a highly lethal cardiovascular condition,making early diagnosis crucial for *** is employed to identify coronary heart disease(CHD).However,due to issues such as fuzzy object boundaries,complex tissue structures,and motion artifacts in ultrasound images,it is challenging to detect CHD *** paper proposes an improved Transformer model based on the Feedback Self-Attention Mechanism(FSAM)for classification of ultrasound *** model enhances attention weights,making it easier to capture complex *** results show that the proposed method achieves high levels of accuracy,recall,precision,F1 score,and area under the receiver operating characteristic curve(72.3%,79.5%,82.0%,81.0%,and 0.73%,respectively).The proposed model was compared with widely used models,including convolutional neural network and visual Transformer model,and the results show that our model outperforms others in the above evaluation *** conclusion,the proposed model provides a promising approach for diagnosing CHD using echocardiogram.

关键词： computer-aided diagnosis(CAD) transformer coronary heart disease feedback self-attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Neighborhood Feature Enhancement Flow Diffusion Model for Point Cloud Generation 27th

Neighborhood Feature Enhancement Flow Diffusion Model for Po...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wang, Hongcheng Zhang, Dongdong Liu, Taotao Qi, Xumai Department of Computer Science and Technology Tongji University Shanghai China

ISBN: (纸本)9783031783883

Collecting 3D point cloud data is cumbersome, so generating high-quality point clouds from existing data can save time and resources while providing more data to support tasks in various fields. In this paper, we propose a neighborhood feature enhancement flow diffusion model for point cloud generation. First, we constructed a multi-scale neighborhood feature aggregation module, which utilizes k-nearest neighbors sampling at different scales to obtain the neighborhood coordinates of each point, thereby aggregating them into coarse global features. Second, we develop a neighborhood attention-based feature enhancement module that uses geometric information in the neighborhood coordinate space to enhance coarse features in the feature space. Then, we used a point-voxel convolutional neural network to reduce redundant features in the enhancement features and output the latent vector of the point cloud. Finally, we transform the latent vectors into data-consistent prior flow features using our designed feature-to-flow data transformation module, seamlessly integrating them into the denoising diffusion model for accurate generation from noisy point clouds. This prior flow approach improves the consistency and coherence of point cloud density distribution. Extensive experiments on the ShapeNet dataset validate the effectiveness of the model in generating 3D point clouds. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Particle Swarm Optimization and Hill Climbing Algorithm for Task Scheduling on Heterogeneous Multicore Clusters 1

引用

3rd International Conference on Advances in Computing Research, ACR 2025

作者： Dietze, Robert Berger, Martin Department of Computer Science Chemnitz University of Technology Chemnitz Germany

ISBN: (数字)9783031876479

ISBN: (纸本)9783031876462

Complex scientific applications often comprise independent tasks which can be executed in parallel on one or more multicore systems. Task scheduling is crucial for the efficient execution of such applications, especially on heterogeneous parallel systems. Since the scheduling of independent tasks is an NP-hard problem, many heuristic and metaheuristic scheduling algorithms have been proposed. Some metaheuristics such as Particle Swarm Optimization (PSO) traverse a wider area of the search space whereas others such as Hill Climbing focus on local optimization. Thus, a promising approach for solving task scheduling problems is the hybridization of PSO and Hill Climbing to exploit the strengths of both methods. In this article, two hybrid variants of PSO and Hill Climbing algorithm for the scheduling of independent tasks to heterogeneous multicore clusters are proposed. In runtime experiments with various types of tasks on a real multicore cluster, both hybrid variants are compared to the single methods regarding schedule execution time and the compute time of the scheduling methods themselves. The results show that the hybridization reduces the schedule execution time while increasing the compute time only slightly. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Fine-Tuning Pre-trained Model with Prototype Learning for OOD Detection 4th

Fine-Tuning Pre-trained Model with Prototype Learning for ...

引用

4th International Conference on Pattern Recognition and Artificial Intelligence, ICPRAI 2024

作者： Liu, Jia-Yi Deng, Zhi-Dong Department of Computer Science and Technology Tsinghua University Beijing China

ISBN: (纸本)9789819787012

Out-of-distribution (OOD) detection primarily learns the in-distribution (ID) representation to distinguish OOD samples from ID classes. Recent methods often fine-tune a pre-trained model with classification objective (cross-entropy, CE) to learn ID representation, disregarding the discrimination between ID and unknown OOD. In this work, we propose to fine-tune pre-trained models in the framework of open-set recognition, which trains a multi-class classifier with ID samples while considering OOD rejection according to distances to ID classes. We adopt prototype learning, which aims to discriminate the ID classes while guiding compact distributions for ID classes in the feature space. The compact ID distributions favor classification and OOD detection using distances, including Euclidean distance and Mahalanobis distance. We evaluated the method on various backbone architectures including ResNet, ViTs and BEiT, and the results show that the prototype learning method on pre-trained models outperforms classification-based method and produces state-of-the-art performance in OOD detection. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Formally Specifying Contract Optimizations with Bisimulations in Coq 6

Formally Specifying Contract Optimizations with Bisimulation...

引用

6th International Workshop on Formal Methods for Blockchains, FMBC 2025

作者： Sorensen, Derek Department of Computer Science and Technology University of Cambridge United Kingdom

ISBN: (纸本)9783959773713

The efficacy of formal verification of smart contracts depends on being able to correctly specify and carry out the verification of optimized code. However, code optimized for performance is rarely optimized for intelligibility, which can make formally verifying optimized code difficult and costly. To address this issue, we present a formal tool for reasoning about an optimized contract in terms of its reference implementation. This tool reduces the work of formally verifying an optimized contract to proving behavioral equivalence to the reference implementation. © Derek Sorensen.

关键词： formal methods interactive theorem prover smart contract upgradeability smart contract verification

来源：评论

学校读者我要写书评

暂无评论

Intelligent Machine Learning Enabled Retinal Blood Vessel Segmentation and Classification

引用

computers, Materials & Continua 2023年第1期74卷 399-414页

作者： Nora Abdullah Alkhaldi Hanan T.Halawani Department of Computer Science College of Computer Science and Information TechnologyKing Faisal UniversityP.O.Box 400AlAhsa31982Saudi Arabia College of Computer Science and Information Systems Najran UniversityNajran61441Saudi Arabia

Automated segmentation of blood vessels in retinal fundus images is essential for medical image *** segmentation of retinal vessels is assumed to be essential to the progress of the decision support system for initial analysis and treatment of retinal *** article develops a new Grasshopper Optimization with Fuzzy Edge Detection based Retinal Blood Vessel Segmentation and Classification(GOFED-RBVSC)*** proposed GOFED-RBVSC model initially employs contrast enhancement ***,GOAFED approach is employed to detect the edges in the retinal fundus images in which the use of GOA adjusts the membership *** ORB(Oriented FAST and Rotated BRIEF)feature extractor is exploited to generate feature ***,Improved Conditional Variational Auto Encoder(ICAVE)is utilized for retinal image classification,shows the novelty of the *** performance validation of the GOFEDRBVSC model is tested using benchmark dataset,and the comparative study highlighted the betterment of the GOFED-RBVSC model over the recent approaches.

关键词： Edge detection blood vessel segmentation retinal fundus images image classification deep learning

来源：评论

学校读者我要写书评

暂无评论

DeepVNE: Deep Reinforcement and Graph Convolution Fusion for Virtual Network Embedding 16

DeepVNE: Deep Reinforcement and Graph Convolution Fusion for...

引用

16th International Conference on COMmunication Systems and NETworkS, COMSNETS 2024

作者： Keerthan Kumar, T.G. Kb, Aneesh Siddheshwar, Anagha Marali, Amulya Kamath, Anupama Koolagudi, S.G. Addya, Sourav Kanti National Institute of Technology Department of Computer Science and Engineering Karnataka India Siddaganga Institute of Technology Department of Computer Science and Engineering India Pes University Department of Computer Science and Engineering India

ISBN: (纸本)9798350383119

Network virtualization (NV) plays a crucial role in modern network management. One of the fundamental challenges in NV is allocating physical network (PN) resources to the demands of the virtual network requests (VNRs). This process is known as a virtual network embedding (VNE) and is NP-hard. Most of the existing approaches for VNE are based on heuristic, meta-heuristic, and exact strategies with limited flexibility and the risk of being stuck in local optimal solutions. In this concern, we provide a deep reinforcement learning (DRL) and graph convolution network (GCN) fusion for VNE (DeepVNE) for maximizing the revenue-to-cost ratio. The DeepVNE takes advantage of the power of actor-critic models within the DRL framework to detect network states and provide optimal solutions matched to current conditions. DeepVNE effectively captures the structural dependencies of both VNRs and PN resources by GCNs, allowing better decision-making during the embedding. Considering several features in the agents throughout the training phase improves resource utilization. The experiments show that DeepVNE outperforms the baselines by gaining a 51% acceptance ratio and 28% revenue-to-cost ratio. © 2024 IEEE.

关键词： Graph Convolutional Network Network Virtualization Reinforcement Learning Virtual Network Embedding

来源：评论

学校读者我要写书评

暂无评论

Visual Robustness Benchmark for Visual Question Answering (VQA)

Visual Robustness Benchmark for Visual Question Answering (V...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Ishmam, Md Farhan Tashdeed, Ishmam Saadat, Talukder Asir Ashmafee, Md Hamjajul Mostofa Kamal, Abu Raihan Hossain, Md. Azam Islamic University of Technology Department of Computer Science and Engineering Bangladesh

ISBN: (纸本)9798331510831

Can Visual Question Answering (VQA) systems maintain their performance when deployed in the real world? Or are they susceptible to realistic corruption effects, e.g., image blur, which can be detrimental in sensitive applications such as medical VQA? While linguistic robustness has been thoroughly explored within the VQA literature, there has yet to be any significant work on visual robustness. In this work, we present the first large-scale benchmark to evaluate the visual robustness of VQA models, including multimodal large language models and zero-shot evaluation, and assess the strength of the realistic corruption effects. Additionally, we have designed several robustness evaluation metrics that quantify an aspect of robustness. These sub-metrics can be aggregated into a unified metric and tailored to fit a variety of use cases. The experiments reveal important insights into the relationships between model size, accuracy, and robustness against the visual corruptions. Our benchmark highlights the need for a balanced approach in model development that considers model performance without compromising robustness. © 2025 IEEE.

关键词： common corruptions corruption robustness evaluation metric multimodal multimodal benchmark multimodal evaluation multimodal learning multimodal robustness robustness robustness benchmark robustness evaluation robustness metric vision language vision language evaluation vision language metric vision language robustness visual corruption visual question answering visual robustness vqa vqa benchmark vqa evaluation vqa metric vqa robustness

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：