检索结果-内蒙古大学图书馆

Simulate the Machine Learning Algorithm to Organize the CRAHN Network System

学校读者我要写书评

暂无评论

Simulate the Machine Learning Algorithm to Organize the CRAH...

Advancement in Electronics & Communication engineering (AECE), International Conference on

作者： P Prabakaran L. Chandra Sekhar Reddy LVD Ravikumar Manish Kumar Verma Department Of Computer Applications School of computing sciences Vels Institute of Science Technology and Advanced Studies Chennai India Department of CSE(Data Science) CMR College of Engineering & Technology Medchal Hyderabad India Department of Information Technology Shri Vishnu Engineering College For Women Bhimavaram India Department of Psychology Lovely Professional University Phagwara India

Using available channels in a wireless spectrum, cognitive radio (CR) may automatically adjust transmission settings to optimize radio operational behaviour. To function properly, a CR ad hoc network (CRAHN) has to be dynamically capable construct autonomous and decentralized networks without negatively impacting licensed main user (PU) systems. For this reason, an effective spectrum necessitates a system structure based on artificial intelligence. This research provides a model for network planning, learning, and dynamic configuration that is based on a distributed autonomous CRAHN network system that uses reinforcement learning. The proposed optimization techniques for spectrum sensing, ad hoc network design, and context-aware signal categorization are all derived from the system model and are based on machine learning. The cognitive and detection engines may be used to examine the spectrum utilization and neighbour network status in the immediate area. To adapt to the ever-changing nature of the wireless environment, the suggested policy engine may generate network operating policies, identify policy conflicts, and infer the best course of action. Together with the erudition engine, whereby apply the recommended machine-learning methods, the decision engine arrives at the best possible settings for the CRAHN. In addition, guarantee peaceful cohabitation with surrounding systems to have excellent signal context recognition ability.

关键词：

Position: Explain to Question not to Justify

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Biecek, Przemyslaw Samek, Wojciech MI2.AI University of Warsaw Poland MI2.AI Warsaw University of Technology Poland Department of Artificial Intelligence Fraunhofer Heinrich Hertz Institute Germany Department of Electrical Engineering and Computer Science Technical University of Berlin Germany BIFOLD - Berlin Institute for the Foundations of Learning and Data Germany

Explainable Artificial Intelligence (XAI) is a young but very promising field of research. Unfortunately, the progress in this field is currently slowed down by divergent and incompatible goals. We separate various threads tangled within the area of XAI into two complementary cultures of human/value-oriented explanations (BLUE XAI) and model/validation-oriented explanations (RED XAI). This position paper argues that the area of RED XAI is currently under-explored, i.e., more methods for explainability are desperately needed to question models (e.g., extract knowledge from well-performing models as well as spotting and fixing bugs in faulty models), and the area of RED XAI hides great opportunities and potential for important research necessary to ensure the safety of AI systems. We conclude this paper by presenting promising challenges in this area. Copyright © 2024, The Authors. All rights reserved.

关键词：

AI-Assisted Digital Pathology for Computational distinction of adenocarcinoma vs squamous cell carcinoma

学校读者我要写书评

暂无评论

AI-Assisted Digital Pathology for Computational distinction ...

International Conference on Mechatronics and Machine Vision in Practice (M2VIP)

作者： Vyomesh Jamwal Abzal Nurgazy Amina Izbassar Munawar Hayat Researcher Robotics & Artificial Intelligence LLP (RAIL) Nazarbayev University Astana Kazakhstan Department of Electrical and Computer Engineering School of Engg. & Digital Sc. Nazarbayev University Astana Kazakhstan Department of Data Science and AI Monash Data Futures Institute Monash University Melbourne Australia

Lung diseases owing to their high morbidity and mortality pose a significant burden to the healthcare systems. The evolution in integrated circuit technology and the consequent rise in computational power has paved the way for the realization of digital pathology (DP) tools. These DP tools equipped with recent developments in artificial intelligence (AI) can assist pathologists in improving clinical evaluations and thereby prescribing betterinformed treatments. Previously researchers have proposed such DP tools for the treatments of breast, prostate, and head and neck cancers, however, developing such tools and applying them for the treatment of lung diseases remains a challenge. The applications of AI-based DP tools on radiology images for improved characterization of lung nodules, lung cancer risk stratification, and fibrotic lung disease are well established. However, the subtypes of non-small cell lung cancer (NSCLC), namely, adenocarcinomas (AC) and squamous cell carcinoma (SCC) are still difficult to distinguish, especially in small biopsies. In this article, we discuss potential architectures of convolutional neural networks (CNN) for the construction of an appropriate AI-based DP classification tool in establishing a crisp distinction between adenocarcinoma and squamous cell carcinoma. It is further emphasized that the construction of such a DP tool will help in mitigating the current issues that the pathologists are facing while working with the conventional molecular biomarkers profiling, a process which is both, expensive and requires high technical and interpretative skills.

关键词：

Deep Active Learning for Multi-Label Classification of Remote Sensing Images

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Möllenbrok, Lars Sumbul, Gencer Demir, Begüm The Faculty of Electrical Engineering and Computer Science Technische Universität Berlin Berlin10623 Germany The BIFOLD - Berlin Institute for the Foundations of Learning and Data Berlin10623 Germany

In this letter, we introduce deep active learning (AL) for multi-label image classification (MLC) problems in remote sensing (RS). In particular, we investigate the effectiveness of several AL query functions for MLC of RS images. Unlike the existing AL query functions (which are defined for single-label classification or semantic segmentation problems), each query function in this paper is based on the evaluation of two criteria: i) multi-label uncertainty;and ii) multi-label diversity. The multilabel uncertainty criterion is associated to the confidence of the deep neural networks (DNNs) in correctly assigning multi-labels to each image. To assess this criterion, we investigate three strategies: i) learning multi-label loss ordering;ii) measuring temporal discrepancy of multi-label predictions;and iii) measuring magnitude of approximated gradient embeddings. The multi-label diversity criterion is associated to the selection of a set of images that are as diverse as possible to each other that prevents redundancy among them. To assess this criterion, we exploit a clustering based strategy. We combine each of the above-mentioned uncertainty strategies with the clustering based diversity strategy, resulting in three different query functions. All the considered query functions are introduced for the first time in the framework of MLC problems in RS. Experimental results obtained on two benchmark archives show that these query functions result in the selection of a highly informative set of samples at each iteration of the AL process. © 2022, CC BY.

关键词： Remote sensing

Deep learning for macular fovea detection based on ultra-widefield fundus images 2

学校读者我要写书评

暂无评论

Deep learning for macular fovea detection based on ultra-wid...

2nd International Conference on Electrical, Electronics, and Information engineering, EEIE 2023

作者： Wang, Mini Han Huang, Lina Hou, Guanghui Yang, Jie Xing, Lumin Yuan, Qiting Chong, Kelvin Kam-Lung Lin, Zhiyuan Zeng, Peijin Fang, Xiaoxiao Yao, Xiaoping Li, Qingqian Liu, Jiang Lin, Chen Department of Ophthalmology & Visual Sciences Faculty of Medicine Chinese University of Hong Kong Hong Kong999077 Hong Kong The faculty of Data Science City University of Macau 999078 China Zhuhai People's Hospital Zhuhai Hospital Affiliated with Jinan University Zhuhai519000 China The Department of AI and big data application Zhuhai Institute of Advanced Technology Chinese Academy of Sciences Zhuhai519000 China Perspective Technology Group Zhuhai519000 China Jinan University Affiliated Shenzhen Eye Hospital Shenzhen518000 China Aier Eye Hospital of Zhuhai Zhuhai519000 China College of Artificial Intelligence Chongqing Industry & Trade Polytechnic Chongqing408000 China The First Affiliated Hospital Shandong First Medical University Shandong Provincial Qianfoshan Hospital Jinan250014 China Business School Hong Kong University of Science and Technology Hong Kong999077 Hong Kong Company Ltd. Guangdong Shenzhen518000 China School of Computer Beijing Institute of Technology Zhuhai Zhuhai519000 China Union Hospital of Fujian Medical University Fuzhou353000 China Department of Ophthalmology People's Hospital of Yantian District Shenzhen518000 China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518000 China Department of Ophthalmology Shenzhen People's Hospital Shenzhen518000 China

ISBN: (纸本)9781510672840

Macula fovea detection is a crucial molecular biological prerequisite for screening and diagnosing macular diseases. Without early detection and proper treatment, any abnormality involving the macula may lead to blindness. However, with the ophthalmologist shortage and time-consuming artificial evaluation, neither the accuracy nor effectiveness of the diagnosis process could be guaranteed. In this project, we proposed a light-weighted deep learning model based on ultra-widefield fundus (UWF) images for macula fovea detection tasks. This study collected 2300 ultra-widefield fundus images from Shenzhen Aier Eye Hospital in China. A light-weighted method based on a U-shape network (Unet) and Fully Convolution Network (FCN) approach is implemented on 1800 (before amplifying process) training fundus images, 400 (before amplifying process) validation images, and 100 test images. Three professional ophthalmologists were invited to mark the fovea. A method from the anatomy perspective is investigated. This approach is derived from the spatial relationship between the macula fovea and optic disc center in UWF. A set of parameters of this method is set based on the experience of ophthalmologists and verified to be effective. The ultra-widefield swept-source optical coherence tomography (UWF-OCT) approach is the grounded method. Through a comparison of proposed methods, we conclude that the proposed light-weighted Unet method outperformed other methods on macula fovea detection tasks. © 2024 SPIE.

关键词： Convolution

Hi-SAM: Marrying Segment Anything Model for Hierarchical Text Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ye, Maoyuan Zhang, Jing Liu, Juhua Liu, Chenyu Yin, Baocai Liu, Cong Du, Bo Tao, Dacheng School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China IFLYTEK Research IFLYTEK CO. LTD. China College of Computing & Data Science Nanyang Technological University #32 Block N4 #02a-014 50 Nanyang Avenue 639798 Singapore

The Segment Anything Model (SAM), a profound vision foundation model pretrained on a large-scale dataset, breaks the boundaries of general segmentation and sparks various downstream applications. This paper introduces Hi-SAM, a unified model leveraging SAM for hierarchical text segmentation. Hi-SAM excels in segmentation across four hierarchies, including pixel-level text, word, text-line, and paragraph, while realizing layout analysis as well. Specifically, we first turn SAM into a high-quality pixel-level text segmentation (TS) model through a parameter-efficient fine-tuning approach. We use this TS model to iteratively generate the pixel-level text labels in a semi-automatical manner, unifying labels across the four text hierarchies in the HierText dataset. Subsequently, with these complete labels, we launch the end-to-end trainable Hi-SAM based on the TS architecture with a customized hierarchical mask decoder. During inference, Hi-SAM offers both automatic mask generation (AMG) mode and promptable segmentation (PS) mode. In the AMG mode, Hi-SAM segments pixel-level text foreground masks initially, then samples foreground points for hierarchical text mask generation and achieves layout analysis in passing. As for the PS mode, Hi-SAM provides word, text-line, and paragraph masks with a single point click. Experimental results show the state-of-the-art performance of our TS model: 84.86% fgIOU on Total-Text and 88.96% fgIOU on TextSeg for pixel-level text segmentation. Moreover, compared to the previous specialist for joint hierarchical detection and layout analysis on HierText, Hi-SAM achieves significant improvements: 4.73% PQ and 5.39% F1 on the text-line level, 5.49% PQ and 7.39% F1 on the paragraph level layout analysis, requiring 20× fewer training epochs. The code is available at Hi-SAM. Copyright © 2024, The Authors. All rights reserved.

关键词： Masks

Hyperedge Learning of Unweighted Brain Hypernetwork Based on Line Graph and Fusion Feature Classification

学校读者我要写书评

暂无评论

science Technology and engineering 2024年第21期24卷 9002-9014页

作者： Shangguan, Xue-Kui Huang, Xiao-Yan Wang, Chun-Yan Guo, Hao Software Engineering Division Shanxi Information Industry Technology Research Institute Co. Ltd. Taiyuan030012 China College of Computer Science and Technology College of Data Science Taiyuan University of Technology Jinzhong030600 China

Brain functional hypernetwork is widely used in the classification diagnosis of brain diseases. In the existing studies, the construction of brain functional hypernetwork is extensively studied to improve classification accuracy, but the impact of brain functional hypernetwork topology on classification performance is overlooked. However, it was showed that hyperedge information could make up for the characteristics of hypernetworks, and the transmission between hyperedges is conducive to overall learning. In view of this problem, hyperedge learning of unweighted brain hypernetwork based on line graph was proposed to analyze the influence of hyperedge on the topology and classification performance. Specifically, firstly, based on functional magnetic resonance data, the star extension method was used to construct brain functional hypernetwork. Secondly, line graph theory was introduced to construct the line graph model. Next, hyperedge density was extracted as local feature and nonparametric test was performed to select feature. Then, the graph-based substructure pattern mining algorithm was adopted to extract subgraph and the frequent score feature selection was applied to select discriminate subgraph. Finally, support vector machine was employed to construct classification model. The results show that the classification results of the proposed method are better than those of traditional brain function hypernetwork, reaching 86. 79%. It is concluded that the topological information of the brain function hypernetwork affected the construction of the classification model. In addition, the fusion features obtained based on the line graph model are better than any single type of features, reaching 88. 68%. It is suggested that that for hyperedge topology information extraction, the attribute information of the hyperedge needs to be considered, and the spatial transfer information ability between hyperedges needs to be considered. © 2024 Beijing Kexue Jishu yu Gongcheng

关键词： Neurodegenerative diseases

ViGT: Proposal-free Video Grounding with Learnable Token in Transformer

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Kun Guo, Dan Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Hefei230601 China Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei230601 China Intelligent Interconnected Systems Laboratory of Anhui Province Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China

The video grounding (VG) task aims to locate the queried action or event in an untrimmed video based on rich linguistic descriptions. Existing proposal-free methods are trapped in complex interaction between video and query, overemphasizing cross-modal feature fusion and feature correlation for VG. In this paper, we propose a novel boundary regression paradigm that performs regression token learning in a transformer. Particularly, we present a simple but effective proposal-free framework, namely Video Grounding Transformer (ViGT), which predicts the temporal boundary using a learnable regression token rather than multi-modal or cross-modal features. In ViGT, the benefits of a learnable token are manifested as follows. (1) The token is unrelated to the video or the query and avoids data bias toward the original video and query. (2) The token simultaneously performs global context aggregation from video and query features. First, we employed a sharing feature encoder to project both video and query into a joint feature space before performing cross-modal co-attention (i.e., video-to-query attention and query-to-video attention) to highlight discriminative features in each modality. Furthermore, we concatenated a learnable regression token [REG] with the video and query features as the input of a vision-language transformer. Finally, we utilized the token [REG] to predict the target moment and visual features to constrain the foreground and background probabilities at each timestamp. The proposed ViGT performed well on three public datasets: ANet Captions, TACoS and YouCookII. Extensive ablation studies and qualitative analysis further validated the interpretability of ViGT. Copyright © 2023, The Authors. All rights reserved.

关键词： Regression analysis

Generative Reasoning Integrated Label Noise Robust Deep Image Representation Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Sumbul, Gencer Demir, Begüm The Faculty of Electrical Engineering and Computer Science Technische Universität Berlin Berlin10623 Germany The BIFOLD - Berlin Institute for the Foundations of Learning and Data Berlin10623 Germany

The development of deep learning based image representation learning (IRL) methods has attracted great attention for various image understanding problems. Most of these methods require the availability of a set of high quantity and quality of annotated training images, which can be time-consuming, complex and costly to gather. To reduce labeling costs, crowdsourced data, automatic labeling procedures or citizen science projects can be considered. However, such approaches increase the risk of including label noise in training data. It may result in overfitting on noisy labels when discriminative reasoning is employed as in most of the existing methods. This leads to sub-optimal learning procedures, and thus inaccurate characterization of images. To address this issue, in this paper, we introduce a generative reasoning integrated label noise robust deep representation learning (GRID) approach. The proposed GRID approach aims to model the complementary characteristics of discriminative and generative reasoning for IRL under noisy labels. To this end, we first integrate generative reasoning into discriminative reasoning through a supervised variational autoencoder. This allows the proposed GRID approach to automatically detect training samples with noisy labels. Then, through our label noise robust hybrid representation learning strategy, GRID adjusts the whole learning procedure for IRL of these samples through generative reasoning and that of the other samples through discriminative reasoning. Our approach learns discriminative image representations while preventing interference of noisy labels during training independently from the IRL method being selected. Thus, unlike the existing label noise robust methods, GRID does not depend on the type of annotation, label noise, neural network architecture, loss function or learning task, and thus can be directly utilized for various image understanding problems. Experimental results show the effectiveness of the proposed GR

关键词： Network architecture