检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Alsmeyer, Gerold Cordero, Fernando Dopmeyer, Hannah Institute for Mathematical Stochastics Department of Mathematics and Computer Science University of Münster Orleans-Ring 10 Münster48149 Germany Institute of Mathematics BOKU University Gregor-Mendel-Straße 33/II Vienna1180 Austria Faculty of Technology Bielefeld University Box 100131 Bielefeld33501 Germany

MSC Codes Primary: 60K05, 92D25, Secondary: 60K10, 60K35, 60J70We consider a population with two types of individuals, distinguished by the resources required for reproduction: type-0 (small) individuals need a fractional resource unit of size ϑ ∈ (0, 1), while type-1 (large) individuals require 1 unit. The total available resource per generation is R. To form a new generation, individuals are sampled one by one, and if enough resources remain, they reproduce, adding their offspring to the next generation. The probability of sampling an individual whose offspring is small is ρR(x), where x is the proportion of small individuals in the current generation. This discrete-time stochastic model is called the two-size Wright-Fisher model, where the function ρR can represent mutation and/or frequency-dependent selection. We show that on the evolutionary time scale, i.e. accelerating time by a factor R, the frequency process of type-0 individuals converges to the solution of a Wright-Fisher-type SDE. The drift term of that SDE accounts for the bias introduced by the function ρR and the consumption strategy, the latter also inducing an additional multiplicative factor in the diffusion term. To prove this, the dynamics within each generation are viewed as a renewal process, with the population size corresponding to the first passage time τ(R) above level R. The proof relies on methods from renewal theory, in particular a uniform version of Blackwell’s renewal theorem for binary, non-arithmetic random variables, established via Ε-coupling. Copyright © 2025, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Stating Comparison Score Uncertainty and Verification Decision Confidence Towards Transparent Face Recognition

arXiv

引用

arXiv 2022年

作者： Huber, Marco Terhörst, Philipp Kirchbuchner, Florian Damer, Naser Kuijper, Arjan Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Technical University of Darmstadt Darmstadt Germany Norwegian University of Science and Technology Gjøvik Norway Paderborn University Paderborn Germany

Face Recognition (FR) is increasingly used in critical verification decisions and thus, there is a need for assessing the trustworthiness of such decisions. The confidence of a decision is often based on the overall performance of the model or on the image quality. We propose to propagate model uncertainties to scores and decisions in an effort to increase the transparency of verification decisions. This work presents two contributions. First, we propose an approach to estimate the uncertainty of face comparison scores. Second, we introduce a confidence measure of the system's decision to provide insights into the verification decision. The suitability of the comparison scores uncertainties and the verification decision confidences have been experimentally proven on three face recognition models on two datasets. Copyright © 2022, The Authors. All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Navigating the Generative AI Project Ecosystem with a Focus on Addressing Data Architecture Complexities and Strategic Model Selection for Optimal Outcomes

Navigating the Generative AI Project Ecosystem with a Focus ...

引用

Emerging Trends in Industry 4.0 (ETI 4.0)

作者： Priyanka Gupta I. Vasantha Kumari B Rajalakshmi Ginni Nijhawan Praveen Zahraa Sahib Mazaal Institute of Aeronautical Engineering Dundigal Hyderabad Medchal Malkajgiri District Siddhartha Institute Of Technology And Sciences Hyderabad Telangana Department of Computer Science New Horizon College of Engineering Bangalore Lovely Professional University Phagwara Knowledge Park II Lloyd Institute of Engineering & Technology Greater Noida Uttar Pradesh Collage of Dentist National University of Science and Technology Dhi Qar Iraq

ISBN: (数字)9798350373783

ISBN: (纸本)9798350373790

This article discusses the Generative Artificial Intelligence (AI) project’s complex ecology and how to pick the optimal models and data design. Creative AI in many commercial and academic fields presents new issues and opportunities. Knowing your way around generative AI projects is crucial. Choose models carefully since data design is hard to grasp. This research provides comprehensive guidance on how to address these areas and ensure project success. Data architecture study is the first step towards generative artificial intelligence since it identifies common issues and viable solutions. Next, to determine which generative models—in terms of complexity, scalability, and ethics—are best suited for a range of tasks, we must evaluate and compare many of them. Researchers were able to validate these findings using case studies and expert interviews. The results suggest that an ordered data infrastructure is required for creative AI. This study indicated the need to boost generational artificial intelligence programs by making practitioners and researchers more aware of the challenges. The study argues that further research and development are necessary to keep up with the fast progress of creative artificial intelligence.

关键词： Ethics Technological innovation Generative AI Biological system modeling Software algorithms computer architecture Data models

来源：评论

学校读者我要写书评

暂无评论

An Exhaustive Meta-Analysis on Consumer Retention Forecasting Using Advanced Machine Learning Techniques

An Exhaustive Meta-Analysis on Consumer Retention Forecastin...

引用

Emerging Trends in Industry 4.0 (ETI 4.0)

作者： Sorabh Lakhanpal V Alekhya B Rajalakshmi Nida Khan Irfan Khan Saad Khudhur Mohammed Lovely Professional University Phagwara Institute of Aeronautical Engineering Dundigal Hyderabad Department of Computer Science New Horizon College of Engineering Bangalore DoCSE GyanGanga Institute of Technology and Sciences Jabalpur M.P India Knowledge Park II Lloyd Institute of Engineering & Technology Greater Noida Uttar Pradesh College of Pharmacy National University of Science and Technology Dhi Qar Iraq

ISBN: (数字)9798350373783

ISBN: (纸本)9798350373790

In this meta-analysis, we present a robust approach for consumer retention forecasting by integrating advanced machine learning algorithms. The method comprises three core algorithms: Consumer Behavior Neural Network (CBNN), Retention Decision Trees (RDT), and Long ShortTerm Memory for Retention (LSTM-R). Our objective is to model intricate relationships within consumer behaviors and accurately predict their retention probabilities. Consumer Behavior Neural Network (CBNN): CBNN, a key component, adeptly captures intricate patterns within consumer behavior data. It involves multiple hidden layers with activation functions, enabling accurate prediction of retention probabilities based on input features. Retention Decision Trees (RDT): This method utilizes decision tree algorithms to predict consumer retention decisions. By recursively splitting the dataset based on information gain, RDT efficiently models consumer behavior, as depicted in the flowchart. Information gain is maximized to create a robust predictive model. Long Short-Term Memory for Retention (LSTM-R): LSTM-R, tailored for temporal data in consumer retention prediction, leverages specialized LSTM architecture. It effectively processes sequential consumer interactions, crucial for accurate prediction of consumer retention probabilities. The proposed method integrates these algorithms, leveraging historical consumer data, and conducts a comprehensive meta-analysis to evaluate their predictive performance.

关键词： Consumer behavior Accuracy Machine learning algorithms Neural networks Machine learning Predictive models Prediction algorithms Decision trees Forecasting Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Profit aware Resource Allocation in Fog Computing: A Stackelberg Game Approach

Profit aware Resource Allocation in Fog Computing: A Stackel...

引用

International Conference on computer, Information and Telecommunication Systems (CITS)

作者： Akshita Doad Nitin Gupta Mohammad S. Obaidat Kuldeep Singh Jadon Piyush Rawat Kuie–Fang Hsiao Department of Computer Science and Engineering National Institute of Technology Hamirpur HP India Distinguished Professor at the King Abdullah II School of Information Technology The University of Jordan Amman Jordan School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China The Amity University Noida UP India University of Texas-Permain Basin Odessa Tx USA

Recently, Internet of Things (IoT) based smart systems have led to more latency-sensitive and bandwidth hungry IoT applications. Fog computing as an extension of Cloud computing fulfills those requirements more effectively. However, a key challenge for the fog computing is multiple resources allocation between heterogeneous Data Service Operators (DSOs) and End Users (EUs). The DSOs want to increase their revenues, however, due to lack of efficient resource allocation mechanisms, they often end up compromising Quality of Service (QoS) and violating Service Level Agreements (SLAs). It is very important for them to make a balance among user's expectations, expenses of providers and penalties for not providing services within the deadline. Hence, a profit-aware resource allocation policy using Stackelberg game is proposed in the work where DSO acts as leaders and EU as followers. The DSOs being the leaders, set the prices first, send it to the followers, and then the followers can compute their best demand for resources accordingly. Next, Stackelberg equilibrium for each type of resource is obtained to find the best resource demand strategy for an EU. The simulation results demonstrate the superiority of the proposed algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Domain-Robust Mitotic Figure Detection with StyleGAN

arXiv

引用

arXiv 2021年

作者： Chung, Youjin Cho, Jihoon Park, Jinah Computer Graphics and Visualizations lab School of Computing Korea Advanced Institute of Science and Technology Daejeon Korea Republic of

We propose a new training scheme for domain generalization in mitotic figure detection. Mitotic figures show different characteristics for each scanner. We consider each scanner as a’domain’ and the image distribution specified for each domain as’style’. The goal is to train our network to be robust on scanner types by using various’style’ images. To expand the style variance, we transfer a style of the training image into arbitrary styles, by defining a module based on StarGAN. Our model with the proposed training scheme shows positive performance on MIDOG Preliminary Test-Set containing scanners never seen before. Copyright © 2021, The Authors. All rights reserved.

关键词： Scanning

来源：评论

学校读者我要写书评

暂无评论

FedHZone: Federated Learning Based Hazardous Zone Detection Using Internet of Drone Things

FedHZone: Federated Learning Based Hazardous Zone Detection ...

引用

Communications, Computing, Cybersecurity, and Informatics (CCCI), International Conference on

作者： Payel Biswas Debashis De Mohammad S. Obaidat Arnab Hazra Balqies Sadoun Department of Computer Science Jogesh Chandra Chaudhuri College Kolkata India Department of Computer Science and Engineering Centre of Mobile Cloud Computing Maulana Abul Kalam Azad University of Technology Nadia WB West Bengal India King Abdullah II School of Information Technology The University of Jordan Amman Jordan School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China Department of Computational Intelligence School of Computing SRM University SRM Nagar Kattankulathur TN India School of Engineering The Amity University Noida UP India Department of Computer Science and Engineering Dr. Sudhir Chandra Sur Institute of Technology & Sports Complex Kolkata India College of Engineering Al Balqa ‘ Applied University Al-Salt Jordan

ISBN: (数字)9798350349832

ISBN: (纸本)9798350349849

Air pollution can affect human health, so it is necessary to predict the air quality index (AQI) in advance. In this work, air quality data collected by the Internet of Drone Things (IoDT) is predicted and analyzed to detect highly polluted zones using a distributed privacy-preserving federated learning (FL) approach. Air quality data for a particular zone is collected and then fed into a Long short-term memory (LSTM) based model to predict the pollution level in local training. Global aggregation is done by an IoDT -mounted central server to aggregate the received models from the local IoDT level. Privacy-preserving FedHZone predicts and analyses the AQI values from the past time series dataset in a multi-drone environment. This work detects the most hazardous zone in a larger region of Jaipur, Rajasthan using real-time data. The result shows FedHZone with stacked LSTM having “'2%-7% better accuracy than the vanilla LSTM-based model.

关键词： Training Federated learning Atmospheric modeling Time series analysis Real-time systems Internet Telecommunication computing Servers Long short term memory Drones

来源：评论

学校读者我要写书评

暂无评论

Literature Survey on Face Recognition of Occluded Faces

Literature Survey on Face Recognition of Occluded Faces

引用

International Conference on Circuit, Power and Computing Technologies (ICCPCT)

作者： Anil J L. Padma Suresh P. Muthukumar S.H. Krishna Veni P. Asha Rajesh Prasad Department of Electrical and Electronics Engineering Nooral Islam Centre for Higher Education Thuckalay Tamil Nadu India Department of Electrical and Electronics Engineering Baselios Mathews II College of Engineering Sasthamcotta Kollam Kerala India Department of Electrical and Electronics Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tiruvallur Chennai Tamilnadu India Department of Computer Science and Engineering Baselios Mathews II College of Engineering Sasthamcotta Kollam Kerala India Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai Tamil Nadu India Department of Electrical and Electronics Engineering Stella Mary’s College of Engineering Tamil Nadu India

ISBN: (数字)9798350372816

ISBN: (纸本)9798350372823

In this paper, a study of the recent advancements in occluded face recognition is made. Recently, this has become very relevant due to the outburst of corona virus. Because of the seriousness of the spread of the disease, every person wore a mask, and it became challenging for automated systems to identify the person. So, it became the need of the hour to find new techniques for occluded face detection. In the years 2021 to 2022, many papers have been published in this regard. Researchers all over the world have contributed significantly towards occluded face recognition. Also, the advancement of deep learning has helped researchers a lot in implementing innovative techniques to deal with the challenges in face recognition, especially occluded faces, which is considered one of the most difficult challenges to overcome. This paper intends to push the limits further and encourage researchers to find more innovative techniques for occlusion-independent face recognition. In this paper, the recent publications in this field are studied. This paper gives an insight into different algorithms which use Convolutional Neural Networks, Deep Learning, Attention mechanism, Dictionary representation, Simultaneous segmentation, Joint segmentation and identification, etc., for occluded face detection and recognition. Incorporating these strategies in face recognition has substantially improved the recognition rate, specifically for occluded face recognition. Additionally, the databases for training occluded face recognition algorithms lack adequate data. This paper also gives an insight into the recent databases available for training occluded face recognition algorithms. Also, different methods for obtaining synthesised occluded faces are discussed.

关键词： Training Deep learning Surveys Representation learning Attention mechanisms Databases Face recognition

来源：评论

学校读者我要写书评

暂无评论

AIR-DA: Adversarial Image Reconstruction for Unsupervised Domain Adaptive Object Detection

arXiv

引用

arXiv 2023年

作者： Sun, Kunyang Lin, Wei Shi, Haoqin Zhang, Zhengming Huang, Yongming Bischof, Horst The School of Information Science and Engineering The National Mobile Communications Research Laboratory Southeast University Nanjing210096 China The Pervasive Communications Center Purple Mountain Laboratories Nanjing211111 China The Institute for Computer Graphics and Vision Graz University of Technology Graz8010 Austria

Unsupervised domain adaptive object detection is a challenging vision task where object detectors are adapted from a label-rich source domain to an unlabeled target domain. Recent advances prove the efficacy of the adversarial based domain alignment where the adversarial training between the feature extractor and domain discriminator results in domain-invariance in the feature space. However, due to the domain shift, domain discrimination, especially on low-level features, is an easy task. This results in an imbalance of the adversarial training between the domain discriminator and the feature extractor. In this work, we achieve a better domain alignment by introducing an auxiliary regularization task to improve the training balance. Specifically, we propose Adversarial Image Reconstruction (AIR) as the regularizer to facilitate the adversarial training of the feature extractor. We further design a multi-level feature alignment module to enhance the adaptation performance. Our evaluations across several datasets of challenging domain shifts demonstrate that the proposed method outperforms all previous methods, of both one- and two-stage, in most settings. Copyright © 2023, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

An Offline Assistance Tool for Visually Impaired People Based on Image Captioning

An Offline Assistance Tool for Visually Impaired People Base...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Guo, Yu Chen, Yue Xie, Yuanyan Ban, Xiaojuan Obaidat, Mohammad S. University of Science and Technology Beijing School of Computer and Communication Engineering Beijing China University of Science and Technology Beijing Shunde Innovation School Foshan China School of Intelligence Science and Technology University of Science and Technology Beijing Beijing China Institute of Artificial Intelligence University of Science and Technology Beijing Beijing China Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing China King Abdullah Ii School of Information Technology University of Jordan Amman Jordan Amity University UP Noida India

ISBN: (纸本)9781665468190

Eye defects of visually impaired people bring great inconvenience to daily lives. With the rapid development of computers and artificial intelligence, some assistance tools and wearable devices have provided convenience to a certain extent, but they still possess many shortcomings, such as high prices, limited functionality, poor real-time performance, the inability to be used offline and so on. This paper designs an offline assistance tool specifically for visually impaired people that can help them perceive their surrounding environments by taking photos and reading the descriptions output by an image captioning model. Considering that the photos taken by visually impaired people may be incomplete, an improved image stitching method based on a genetic algorithm is proposed to obtain high-quality stitched images as the inputs of the image captioning model. An improved model pruning algorithm is also designed to compress and accelerate the image captioning model to meet the offline and real-time deployment requirements of portable devices. Experimental results and a practical tool show that the developed system based on an image captioning model and pruning can quickly and accurately describe environments, thus assisting visually impaired people anytime and anywhere. © 2022 IEEE.

关键词： Genetic algorithms

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：