检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Diakonikolas, Jelena Guzmán, Cristóbal Department of Computer Sciences University of Wisconsin-Madison United States Institute for Mathematical and Computational Engineering Faculty of Mathematics School of Engineering Pontificia Universidad Católica de Chile Chile

We initiate the study of nonsmooth optimization problems under bounded local subgradient variation, which postulates bounded difference between (sub)gradients in small local regions around points, in either average or maximum sense. The resulting class of objective functions encapsulates the classes of objective functions traditionally studied in the optimization literature, which are defined based on either Lipschitz continuity of the objective or Hölder/Lipschitz continuity of the function’s gradient. Further, the defined class is richer in the sense that it contains functions that are neither Lipschitz continuous nor have a Hölder continuous gradient. Finally, when restricted to the aforementioned traditional classes of optimization problems, the constants defining the studied classes lead to more fine-grained oracle complexity bounds. Some highlights of our results are that: (i) it is possible to obtain complexity results for both convex and nonconvex optimization problems with (local or global) Lipschitz constant being replaced by a constant of local subgradient variation, corresponding to small local regions and (ii) complexity of the subgradient set around the set of optima – measured by its mean width in a local region around optima – plays a role in the complexity of nonsmooth optimization, particularly in parallel optimization settings. A consequence of (ii) is that for any error parameter ϵ > 0, parallel oracle complexity of nonsmooth Lipschitz convex optimization is lower than its sequential oracle complexity by a factor Ω˜(1ϵ) whenever the objective function is piecewise-linear with the number of pieces polynomial in the dimension and 1/ϵ. This is particularly surprising considering that existing parallel complexity lower bounds are based on such classes of functions. The seeming contradiction is resolved by considering the region in which the algorithm is allowed to query the objective. © 2024, CC BY.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

PolluSenseCheck: Cost Effective TinyML-based Air Quality Monitoring System with In-Built Sensor Fault Detection

PolluSenseCheck: Cost Effective TinyML-based Air Quality Mon...

引用

International Communication Systems and Networks and Workshops, COMSNETS

作者： Arko Datta Joyita Chakraborty Pabitra Majhi Astom Mondal Satyaki Roy Subrata Nandi Sujoy Saha Department of Computer Science and Engineering National Institute of Technology Durgapur India Department of Computer Science and Engineering Dr. B. C. Roy Engineering College Durgapur India Department of Mathematical Sciences University of Alabama in Huntsville Alabama USA

ISBN: (数字)9798331531195

ISBN: (纸本)9798331531201

Air pollution is a significant contributor to respiratory diseases, posing severe health risks globally. Traditional air quality monitoring stations, while effective, are too expensive, particularly in developing nations like India, where numerous monitoring sites may be required to cover diverse urban and industrial regions. Existing cost-effective air quality monitoring systems are rendered inadequate by their reliance on sensors that are prone to drift, which necessitates frequent recalibration. This study addresses the gap by developing a cost-effective air quality monitoring device capable of detecting particulate matter, carbon monoxide (CO), temperature, and humidity, and incorporating self-diagnostic capabilities. It enhances reliability through a dual-sensing mechanism: it calculates the Air Quality Index (AQI) directly and uses a TinyML model to provide secondary-sensed AQI estimations. Deviations between these two AQI values signal sensor malfunction, which is tracked by a proposed algorithm, Interocept, that assigns fault scores to the sensors. Models were created for eight cities, showing accuracies from 0.69 – 0.89 for dust and 0.76 – 0.95 for CO. These models were optimized for real-time performance by converting them into TensorFlow Lite micro-models, allowing deployment on the device itself for continuous, autonomous sensor health diagnostics and improved long-term air quality management systems.

关键词： Temperature sensors Atmospheric modeling Tiny machine learning Urban areas Air quality Sensor systems Real-time systems Sensors Reliability Monitoring

来源：评论

学校读者我要写书评

暂无评论

Credit vs. Discount-Based Congestion Pricing: A Comparison Study

arXiv

引用

arXiv 2024年

作者： Chiu, Chih-Yuan Jalota, Devansh Pavone, Marco University of California Berkeley Department of Electrical Engineering and Computer Sciences United States Stanford University Institute for Computational & Mathematical Engineering United States Stanford University Department of Aeronautics and Astronautics United States

Congestion pricing offers a promising traffic management policy for regulating congestion, but has also been criticized for placing outsized financial burdens on low-income users. Credit-based congestion pricing (CBCP) and discount-based congestion pricing (DBCP) policies, which respectively provide travel credits and toll discounts to subsidize low-income users’ access to tolled roads, are promising mechanisms for reducing traffic congestion without worsening societal inequities. However, the optimal design and relative merits of CBCP and DBCP policies remain poorly understood. This work studies the effects of deploying CBCP and DBCP policies to route users on multi-lane highway networks with tolled express lanes. We formulate a non-atomic routing game in which a subset of eligible users is granted toll relief via a fixed budget or toll discount, while the remaining ineligible users must pay out-of-pocket. We prove that Nash equilibrium traffic flow patterns exist under any CBCP or DBCP policy. Under the additional assumption that eligible users have time-invariant values of time (VoTs), we provide a convex program to efficiently compute these equilibria. Moreover, for single-edge networks, we identify conditions under which DBCP policies outperform CBCP policies, in the sense of improving eligible users’ express lane access, an equity objective often neglected in existing congestion pricing schemes that price low-income users out of utilizing express lanes. Specifically, we identify user and network-dependent parameters whose values play a key role in determining whether DBCP or CBCP policies are more effective at expanding eligible users’ express lane access. Finally, we present empirical results from a CBCP pilot study of the San Mateo 101 Express Lane Project in California. Our empirical results corroborate our theoretical analysis of the impact of deploying credit-based and discount-based policies, and lend insights into the sensitivity of their impact with re

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Quantum-Accelerated Hyperparameter Tuning for Dynamic NLP Models

引用

Procedia computer science 2024年 235卷 506-519页

作者： Ravikumar S Arockia Raj Y R. Babu Vijay K R. Ramani Associate Professor Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Tamilnadu India 600062 Assistant Professor Department of Computer Science and Engineering PSNA College of Engineering and Technology Dindigul India Assistant Professor Department of Computational Intelligence Faculty of College of Engineering and Technology SRM Institute of Science and Technology Chennai Tamilnadu India 603203 Assistant Professor Department of Computer Science and Engineering Rajalakshmi Engineering College Chennai Tamilnadu India 602117 Associate Professor Department of Computer Science and Engineering P.S.R. Engineering College Sivakasi Tamilnadu India 626140

In the rapidly evolving field of natural language processing (NLP), performance optimization of large-scale NLP models is crucial. Through the application of Quantum-Accelerated Hyperparameter Tuning (QAHT), this abstract introduces a novel approach to addressing this issue. Our proposed framework leverages quantum computing capabilities to dynamically optimize NLP model hyperparameters in real-time, catering to the ever-changing character of textual data streams. Traditional hyperparameter optimisation methods usually rely on laborious grid searches or random exploration, which may not be suitable for dynamic NLP jobs. Contrarily, QAHT uses Quantum Neural Network (QNN) architectures that have been specially designed for hyperparameter optimisation. These QNNs improve performance and efficacy by dynamically modifying and improving model configurations. This abstract discusses the key elements of the QAHT architecture, including real-time model deployment, adaptive learning, and continuous data stream processing. In addition to speeding up the hyperparameter optimisation process, QAHT makes sure that NLP models are still flexible and responsive to shifts in the sentiment of the data and its distribution. This method has applications beyond NLP since it provides a foundation for effectively optimising machine learning models in complex, real-time situations. As quantum computing develops, QAHT represents a promising future in machine learning, where quantum-enhanced capabilities satisfy the needs of contemporary data-driven applications.

关键词： Hyperparameter optimization Quantum computing Natural language processing (NLP) Quantum Neural Networks (QNN) Real-time processing Adaptive learning

来源：评论

学校读者我要写书评

暂无评论

Patient-specific coronary angioplasty simulations — A mixed-dimensional finite element modeling approach

引用

computers in Biology and Medicine 2025年 189卷 109914-109914页

作者： Datz, Janina C. Steinbrecher, Ivo Meier, Christoph Hagmeyer, Nora Engel, Leif-Christopher Popp, Alexander Pfaller, Martin R. Schunkert, Heribert Wall, Wolfgang A. Institute for Computational Mechanics Technical University of Munich Germany Department of Cardiology Deutsches Herzzentrum München Technical University of Munich Germany Institute for Mathematics and Computer-Based Simulation University of the Bundeswehr Munich Germany Pediatric Cardiology Cardiovascular Institute and Institute for Computational and Mathematical Engineering Stanford University United States Munich Institute of Biomedical Engineering Technical University of Munich Germany

Coronary angioplasty with stent implantation is the most frequently used interventional treatment for coronary artery disease. However, reocclusion within the stent, referred to as in-stent restenosis, occurs in up to 10% of lesions. It is widely accepted that mechanical loads on the vessel wall strongly affect adaptive and maladaptive mechanisms. Yet, the role of procedural and lesion-specific influence on restenosis risk remains understudied. computational modeling of the stenting procedure can provide new mechanistic insights, such as local stresses, that play a significant role in tissue growth and remodeling. Previous simulation studies often featured simplified artery and stent geometries and cannot be applied to real-world examples. Realistic simulations were computationally expensive since they featured fully resolved stenting device models. The aim of this work is to develop and present a mixed-dimensional formulation to simulate the patient-specific stenting procedure with a reduced-dimensional beam model for the stent and 3D models for the artery. In addition to presenting the numerical approach, we apply it to realistic cases to study the intervention's mechanical effect on the artery and correlate the findings with potential high-risk locations for in-stent restenosis. We found that high artery wall stresses develop during the coronary intervention in severely stenosed areas and at the stent boundaries. Herewith, we lay the groundwork for further studies towards preventing in-stent restenosis after coronary angioplasty. © 2025 The Authors

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

Handling Imbalanced Data for Credit Card Fraudulent Detection: A Machine Learning Approach 1st

Handling Imbalanced Data for Credit Card Fraudulent Detectio...

引用

1st International Conference on Advances in Artificial Intelligence and Machine Learning in Big Data Processing, AAIMB 2023

作者： Sujatha, E. Umarani, V. Rekha, K.S. Gopirajan, P.V. Manickavasagan, V. Sriperumbudur India Department of Computational Intelligence SRM Institute of Science and Technology Chennai India Department of Computer Science and Business Systems Sri Sairam Engineering College Chennai India

ISBN: (纸本)9783031730672

Imbalanced data is crucial for contemporary machine learning models. Evaluating the effectiveness of machine learning classifiers under an unfair dataset is problematic. Several resampling techniques were applied to the unwarranted dataset and pre-processing stages to increase the accuracy of prognosis classifiers. This proposed work discusses the basics of resampling techniques such as Random under-sampling, Random oversampling, SMOTE (Synthetic Minority Over-sampling Technique), SMOTE-ENN (Synthetic minority over-sampling technique and Edited nearest neighbor), ADASYN (Adaptive Synthetic), and SMOTE-Tomek to handle this Weakness with unbalanced data. The proposed system uses an unstable credit card fraudulent dataset collected from Kaggle, which contains 21 columns and 28382 rows, of which 18497 negative samples and 4208 positive samples. The balanced dataset got by applying resampling methods SMOTE, SMOTE-ENN, ADASYN, and SMOTE-Tomek. Afterward, it uses machine learning techniques such as Logistic Regression (LR), Decision tree (DT), and Random Forest (RF) for classification. From this analysis, it selects the optimal scheme and classifier. The grey wolf optimization technique improves the classifier’s performance, which optimizes the samples after sampling. This hybrid method’s effectiveness was evaluated with standard evaluation parameters. The hybrid Random Forest technique achieved 92% accuracy from this analysis compared to hybrid models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Hybrid Deep Learning Models with Genetic Algorithm Optimization for Enhanced Kidney Tumor Detection

Hybrid Deep Learning Models with Genetic Algorithm Optimizat...

引用

Electronics and Sustainable Communication Systems (ICESC), 2020 International Conference on

作者： Rajagopal K V. Sheeja Kumari Saraswathy S V. Suresh Kumar S. Ponmaniraj A. Deepa Computer Science and Engineering Saveetha Institute of Medical and Technical Sciences SIMATS University Thandalam Chennai Department of Computer Science and Engineering Saveetha Institute of Medical and Technical Sciences SIMATS University Thandalam Chennai Department of Computer Science and Business Systems Sri Sairam Engineering College Chennai Department of Computational Intelligence Institute of ECE Saveetha Institute of Medical and Technical Sciences SIMATS University Thandalam Chennai Department of Computer Science and Engineering Saveetha Engineering College Thandalam Chennai

ISBN: (数字)9798350379945

ISBN: (纸本)9798350379952

Accurate and timely segmentation of kidney tumors from medical images is crucial for effective treatment and prognosis. This research study proposes a novel hybrid deep learning framework incorporating Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for enhanced kidney tumor segmentation. To optimize model performance, Genetic Algorithms (GA) are employed to fine-tune hyperparameters and network architecture. The proposed method effectively captures both local and global image features, leading to improved segmentation accuracy compared to existing techniques. Experimental results on a publicly available dataset demonstrate the superiority of our approach in terms of recall, F1-score, and precision. This research contributes to the advancement of computer-aided diagnosis for kidney cancer.

关键词： Deep learning Image segmentation Recurrent neural networks Accuracy Network architecture Convolutional neural networks Kidney Optimization Tumors Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Resolving Attached USBs: Analysis of Windows 11 Artifacts

Resolving Attached USBs: Analysis of Windows 11 Artifacts

引用

Advancement in Technology (ICONAT), International Conference for

作者： Akash Budhrani Upasna Singh Bhupendra Singh School of Computer Engineering and Mathematical Science Defence Institute of Advanced Technology Pune India Computer Science and Engineering Department Indian Institute of Information Technology Pune India

The Windows Operating System is known for its convenience which tends to breed more and more user information in form of Artifacts. Artifacts are important repository of potential evidence while conducting any computer investigation. This paper thus provides an analysis of artifacts that will bring out the inside of USB devices attached or mounted on a particular system. Moreover, this paper will parse the data out into usable information and helps to interpret the findings of forensic examiner. In this work, Image file of the suspected system has been exported from the system for further analysis to bring out relevant information.

关键词： Image resolution Operating systems Forensics Universal Serial Bus

来源：评论

学校读者我要写书评

暂无评论

Dimensionality and dynamics for next-generation artificial neural networks

Patterns

引用

Patterns 2025年

作者： Wang, Ge Fan, Feng-Lei Department of Biomedical Engineering Department of Electrical Computer and Systems Engineering Department of Computer Science Center for Computational Innovations Biomedical Imaging Center Center for Biotechnology and Interdisciplinary Studies Rensselaer Polytechnic Institute TroyNY United States Department of Data Science City University of Hong Kong Kowloon Hong Kong

The recent awarding of the Nobel Prize in Physics to Geoffrey E. Hinton and John J. Hopfield highlights their profound impact on artificial neural networks. In this perspective, we explore how their foundational insights can drive the advancement of next-generation artificial intelligence (AI) models. We propose expanding beyond conventional architectures by introducing dimensionality through intra-layer links and dynamics via feedback loops. Network height and additional dimensions, alongside traditional width and depth, enhance learning capabilities, while entangled loops across scales induce emergent behaviors akin to phase transitions in physics. We discuss how these principles extend beyond transformers, fostering a new paradigm of intelligence inspired by physics-driven models and biological cognition mechanisms. © 2025 The Author(s)

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Improving the Detection of Kidney Tumours Using Firefly Algorithm Optimisation

Improving the Detection of Kidney Tumours Using Firefly Algo...

引用

Intelligent Cyber Physical Systems and Internet of Things (ICoICI), International Conference on

作者： Rajagopal K V. Sheeja Kumari Saraswathy S S. Ponmaniraj V. Suresh Kumar A. Deepa Department of Computer Science and Engineering Saveetha Institute of Medical and Technical Sciences SIMATS University Thandalam Chennai Department of Computer Science and Business Systems Sri Sairam Engineering College Chennai Department of Computational Intelligence Institute of ECE Saveetha Institute of Medical and Technical Sciences SIMATS University Thandalam Chennai Department of Computer Science and Engineering Saveetha Engineering College Thandalam Chennai

ISBN: (数字)9798331540661

ISBN: (纸本)9798331540678

Accurate and timely diagnosis of kidney tumors is crucial for effective treatment. This study proposes a novel approach utilizing deep learning techniques and the firefly algorithm (FA) to enhance kidney tumor segmentation. By optimizing feature selection and CNN hyperparameters, FA improves the accuracy and efficiency of the detection system. The proposed method involves pre-processing medical images, extracting features using CNNs, and fine-tuning the model with FA. Experiments on a public dataset demonstrate significant improvements in classification metrics compared to traditional methods. The FA-optimized model provides clinicians with a valuable tool for accurate kidney tumor detection and classification, leading to improved patient outcomes.

关键词： Deep learning Image segmentation Accuracy Feature extraction Classification algorithms Convolutional neural networks Kidney Optimization Medical diagnostic imaging Tumors

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：