检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Nuoya Xiong Zhaoran Wang Zhuoran Yang IIIS Tsinghua University China Department of Industrial Engineering and Management Sciences Northwestern University Department of Statistics and Data Science Yale University

We take the first step in studying general sequential decision-making under two adaptivity constraints: rare policy switch and batch learning. First, we provide a general class called the Eluder Condition class, which includes a wide range of reinforcement learning classes. Then, for the rare policy switch constraint, we provide a generic algorithm to achieve a Õ(log K) switching cost with a Õ(log √K) regret on the EC class. For the batch learning constraint, we provide an algorithm that provides a Õ(√K + K/B) regret with the number of batches B. This paper is the first work considering rare policy switch and batch learning under general function classes, which covers nearly all the models studied in the previous works such as tabular MDP (Bai et al., 2019; Zhang et al., 2020), linear MDP (Wang et al., 2021; Gao et al., 2021), low eluder dimension MDP (Kong et al., 2021; Velegkas et al., 2022), generalized linear function approximation (Qiao et al., 2023), and also some new classes such as the low DΔ-type Bellman eluder dimension problem, linear mixture MDP, kernelized nonlinear regulator and undercomplete partially observed Markov decision process (POMDP).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalized equivalences between subsampling and ridge regularization 23

Generalized equivalences between subsampling and ridge regul...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Pratik Patil Jin-Hong Du Department of Statistics University of California Berkeley CA Department of Statistics and Data Science & Machine Learning Department Carnegie Mellon University Pittsburgh PA

We establish precise structural and risk equivalences between subsampling and ridge regularization for ensemble ridge estimators. Specifically, we prove that linear and quadratic functionals of subsample ridge estimators, when fitted with different ridge regularization levels λ and subsample aspect ratios ψ, are asymptotically equivalent along specific paths in the (λ, ψ)-plane (where ψ is the ratio of the feature dimension to the subsample size). Our results only require bounded moment assumptions on feature and response distributions and allow for arbitrary joint distributions. Furthermore, we provide a data-dependent method to determine the equivalent paths of (λ, ψ). An indirect implication of our equivalences is that optimally tuned ridge regression exhibits a monotonic prediction risk in the data aspect ratio. This resolves a recent open problem raised by Nakkiran et al. [1] for general data distributions under proportional asymptotics, assuming a mild regularity condition that maintains regression hardness through linearized signal-to-noise ratios.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Holistic Anti-Counterfeiting Platform Using NFC and Blockchain Technologies

引用

computers, Materials & Continua 2025年第6期83卷 4257-4280页

作者： Rajendren Subramaniam Saaidal Razalli Azzuhri Teh Ying Wah Atif Mahmood Vimala Balakrishnan Department of Information System Faculty of Computer Science&Information TechnologyUniversiti MalayaKuala Lumpur50603Malaysia Department of Computer System&Technology Faculty of Computer Science&Information TechnologyUniversiti MalayaKuala Lumpur50603Malaysia Faculty of Data Science and Information Technology INTI International UniversityNilai71800Malaysia

Counterfeiting is still a pervasive global issue,affecting multiple industries and hindering industrial innovation,while causing substantial financial losses,reputational damage,and risks to consumer *** luxury goods and pharmaceuticals to electronics and automotive parts,counterfeit products infiltrate supply chains,leading to a loss of revenue for legitimate businesses and undermining consumer *** anti-counterfeiting measures,such as holograms,serial numbers,and barcodes,have proven to be insufficient as counterfeiters continuously develop more sophisticated replication *** a result,there is a growing need for more advanced,secure,and reliable methods to prevent *** paper presents a novel,holistic anti-counterfeiting platform that integrates Near Field Communication(NFC)-enabled mobile applications with blockchain technology to provide an innovative,secure,and consumer-friendly authentication *** approach addresses key gaps in existing solutions by incorporating dynamic product identifiers,which make replication significantly more *** system enables consumers to verify the authenticity of products instantly using their smartphones,enhancing transparency and trust in the supply *** technology plays a crucial role in our proposed solution by providing an immutable,decentralized ledger that records product authentication *** ensures that product verification records cannot be tampered with or altered,adding a layer of security that is absent in conventional ***,NFC technology enhances security by offering unique identification capabilities,enabling real-time product *** validate the effectiveness of the proposed system,real-world testing was conducted across different *** results demonstrated the platform’s ability to significantly reduce counterfeit products in the supply chain,offering businesses and consumers a more robust and reliable aut

关键词： Blockchain near field communication(NFC) anti-counterfeiting

来源：评论

学校读者我要写书评

暂无评论

Establishment of an Effective Brain Tumor Classification System through Image Transformations and Optimization Techniques 1

Establishment of an Effective Brain Tumor Classification Sys...

引用

1st IEEE International Conference on Innovations in High Speed Communication and Signal Processing, IHCSP 2023

作者： Arora, Gaurav Kumar Taj Mahaboob, Shaik Adilakshmi, S. Rani, Sasi Kala Kasthuri, A. Mamodiya, Udit IBM India Pvt. Ltd Uttar Pradesh Greater Noida India Jntua College of Engineering Department of ECE Andhra Pradesh Pulivendula India Vignana Bharathi Institute of Technology Department of Data Science Telangana India Sri Krishna College of Engineering and Technology Department Computer Science and Engineering Coimbatore India Vinayaka Mission's Kirupananda Variyar Engineering College Department Computer Science and Engineering Salem India Poornima Institute of Engineering and Technology Department of Electrical Engineering Rajasthan Jaipur India

ISBN: (纸本)9798350345957

In recent years, researchers and health care professionals have been interested in medical image fusion, which combines images of the human body, organs, and cells with general information to address medical difficulties. computer-aided imaging techniques offer quantitative assessment of images under validation, which helps doctors make unbiased, objective decisions quickly. Multi-sensor and multi-source image fusion models also diversify healthcare analysis characteristics. It led to strong information processing that reveals otherwise invisible details. The fused pictures can help pinpoint anomalies. Multimodal image fusion combines complementing pictures into one. The fused image is ideal for visual perception or computer segmentation, feature extraction, and object recognition. The goal is to create a single fused image with more information than the source photos alone. The fused image combines the source images' complementary and shared elements, delivering enhanced subjective and objective detail. In biomedical imaging, fusing multimodal images has shown useful for disease characterisation. Medical image analysis uses CT, MRI, and PET (PET). In this research paper, we developed a Multimodality Image Fusion using Centre based genetic algorithm (CBGA) and fuzzy logic (FL) technique with lifting wavelet transform (LWT) for fusion of spatially registered MRI and CT images. We also proposed an efficient Multimodality Medical Image Fusion system based on DWT and binary crow search optimization (BCSO) algorithm for fusing MRI and CT images. The performance of the suggested models is validated against many benchmark medical datasets. The experiment showed that the proposed models outperformed all others. © 2023 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Deep Learning based Smart American Sign Language Detection in Alphabets 6

Deep Learning based Smart American Sign Language Detection i...

引用

6th International Conference on Contemporary Computing and Informatics, IC3I 2023

作者： RukumaniKhandhan, C. Vijayakumar, P. Sri Sabari, C. Arun, G. Chandrasekaran, N. Rohith Bhat, C. Kongu Engineering College Department of Computer Technology - PG Erode Perundurai India Kongu Engineering College Department of Master of Computer Applications Tamil Nadu Perundurai India Kongu Engineering College Department of Computer Technology - PG Tamil Nadu Perundurai India Erode Sengunthar Engineering College Department of Artificial Intelligence and Data Science Thuduppathi India PSNA College of Engineering and Technology Department of Electrical and Electronics Engineering Dindigul India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350304480

Disabled people are of different kinds like physically challenged people, mentally challenged etc. To help the people who cannot convey their thoughts and who cannot hear others thoughts, sign language is being used to communicate with them. Sign language is something that is used to communicate through body movements mainly through hands and *** Sign Language(ASL) is an authorized sign language that is taught and used widely. ASL is difficult to teach and learn for them because it is completely new and they have no knowledge about it. In this research paper the ASL is recognized and converted into text through data mining techniques. The study uses a dataset of ASL alphabet images and applies various data mining techniques, including feature selection, classification, and clustering. The features extracted from the images include shape, size, color, and texture. The use of neural networks, decision trees, and support vector machines speeds up the data categorization process. Techniques such as k-means clustering and hierarchical grouping are used to facilitate entity grouping. The results show that the best performance is achieved using a combination of feature selection and classification techniques. The decision tree algorithm with the selected features achieves an accuracy of 96.5% on the test dataset. The system can accurately recognize all the ASL alphabets and convert them into text or speech. The system developed can be used to improve communication between deaf and hearing individuals and can be extended to recognize other ASL gestures and phrases. Future research can focus on real-time detection and recognition of ASL gestures using data mining techniques. CNN algorithm is used to recognize the image based on the images collected. The accuracy rate gets low and the detection is inaccurate. Therefore, VGG16 algorithm is used to recognize the images and it is unique in convolutional models. © 2023 IEEE.

关键词： ASL CNN Gestures VGG16

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning for Latent Variable Network Models

arXiv

引用

arXiv 2024年

作者： Jalan, Akhil Mazumdar, Arya Mukherjee, Soumendu Sundar Sarkar, Purnamrita Department of Computer Science UT Austin United States Halıcıoğlu Data Science Institute Dept of CSE UC San Diego United States Indian Statistical Institute Kolkata India Department of Statistics & Data Sciences UT Austin United States

We study transfer learning for estimation in latent variable network models. In our setting, the conditional edge probability matrices given the latent variables are represented by P for the source and Q for the target. We wish to estimate Q given two kinds of data: (1) edge data from a subgraph induced by an o(1) fraction of the nodes of Q, and (2) edge data from all of P. If the source P has no relation to the target Q, the estimation error must be Ω(1). However, we show that if the latent variables are shared, then vanishing error is possible. We give an efficient algorithm that utilizes the ordering of a suitably defined graph distance. Our algorithm achieves o(1) error and does not assume a parametric form on the source or target networks. Next, for the specific case of Stochastic Block Models we prove a minimax lower bound and show that a simple algorithm achieves this rate. Finally, we empirically demonstrate our algorithm’s use on real-world and simulated network estimation problems. © 2024, CC BY-NC-ND.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Value-at-Risk- and Expectile-based Systemic Risk Measures and Second-order Asymptotics: With Applications to Diversification

arXiv

引用

arXiv 2024年

作者： Geng, Bingzhen Liu, Yang Zhao, Yimiao School of Big Data and Statistics Anhui University China School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Department of Statistics and Actuarial Science University of Waterloo Canada

The systemic risk measure plays a crucial role in analyzing individual losses conditioned on extreme system-wide disasters. In this paper, we provide a unified asymptotic treatment for systemic risk measures. First, we classify them into two families of Value-at-Risk- (VaR-) and expectile-based systemic risk measures. While VaR has been extensively studied, in the latter family, we propose two new systemic risk measures named the Individual Conditional Expectile (ICE) and the Systemic Individual Conditional Expectile (SICE), as alternatives to Marginal Expected Shortfall (MES) and Systemic Expected Shortfall (SES). Second, to characterize general mutually dependent and heavy-tailed risks, we adopt a modeling framework where the system, represented by a vector of random loss variables, follows a multivariate Sarmanov distribution with a common marginal exhibiting second-order regular variation. Third, we provide second-order asymptotic results for both families of systemic risk measures. This analytical framework offers a more accurate estimate compared to traditional first-order asymptotics. Through numerical and analytical examples, we demonstrate the superiority of second-order asymptotics in accurately assessing systemic risk. Further, we conduct a comprehensive comparison between VaR-based and expectile-based systemic risk measures. We find that expectile-based measures output higher risk evaluation than VaR-based ones, emphasizing the former's potential advantages in reporting extreme events and tail risk. As a financial application, we use the asymptotic treatment to discuss the diversification benefits associated with systemic risk measures. The financial insight is that the expectile-based diversification benefits consistently deduce an underestimation and suggest a conservative approximation, while the VaR-based diversification benefits consistently deduce an overestimation and suggest behaving optimistically. Copyright © 2024, The Authors. All rights reser

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

Collaborative Pareto Set Learning in Multiple Multi-Objective Optimization Problems

Collaborative Pareto Set Learning in Multiple Multi-Objectiv...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Chikai Shang Rongguang Ye Jiaqi Jiang Fangqing Gu School of Mathematics and Statistics Guangdong University of Technology China Department of Computer Science and Engineering Southern University of Science and Technology China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Pareto Set Learning (PSL) is an emerging research area in multi-objective optimization, focusing on training neural networks to learn the mapping from preference vectors to Pareto optimal solutions. However, existing PSL methods are limited to addressing a single Multi-objective Optimization Problem (MOP) at a time. When faced with multiple MOPs, this limitation results in significant inefficiencies and hinders the ability to exploit potential synergies across varying MOPs. In this paper, we propose a Collaborative Pareto Set Learning (CoPSL) framework, which learns the Pareto sets of multiple MOPs simultaneously in a collaborative manner. CoPSL particularly employs an architecture consisting of shared and MOP-specific layers. The shared layers are designed to capture commonalities among MOPs collaboratively, while the MOP-specific layers tailor these general insights to generate solution sets for individual MOPs. This collaborative approach enables CoPSL to efficiently learn the Pareto sets of multiple MOPs in a single execution while leveraging the potential relationships among various MOPs. To further understand these relationships, we experimentally demonstrate that shareable representations exist among MOPs. Leveraging these shared representations effectively improves the capability to approximate Pareto sets. Extensive experiments underscore the superior efficiency and robustness of CoPSL in approximating Pareto sets compared to state-of-the-art approaches on a variety of synthetic and real-world MOPs. Code is available at https://***/ckshang/CoPSL.

关键词： Training Codes Neural networks Collaboration Focusing Pareto optimization Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Revisiting Active Learning in the Era of Vision Foundation Models

arXiv

引用

arXiv 2024年

作者： Gupte, Sanket Rajan Aklilu, Josiah Nirschl, Jeffrey J. Yeung-Levy, Serena Department of Computer Science Stanford University United States Department of Biomedical Data Science Stanford University United States Department of Pathology Stanford University United States

Foundation vision or vision-language models are trained on large unlabeled or noisy data and learn robust representations that can achieve impressive zero- or few-shot performance on diverse tasks. Given these properties, they are a natural fit for active learning (AL), which aims to maximize labeling efficiency. However, the full potential of foundation models has not been explored in the context of AL, specifically in the low-budget regime. In this work, we evaluate how foundation models influence three critical components of effective AL, namely, 1) initial labeled pool selection, 2) ensuring diverse sampling, and 3) the trade-off between representative and uncertainty sampling. We systematically study how the robust representations of foundation models (DINOv2, OpenCLIP) challenge existing findings in active learning. Our observations inform the principled construction of a new simple and elegant AL strategy that balances uncertainty estimated via dropout with sample diversity. We extensively test our strategy on many challenging image classification benchmarks, including natural images as well as out-of-domain biomedical images that are relatively understudied in the AL literature. We also provide a highly performant and efficient implementation of modern AL strategies (including our method) at https://***/sanketx/AL-foundation-models. © 2024, CC BY.

关键词： Foundations

来源：评论

学校读者我要写书评

暂无评论

Automated Depression Detection from Text and Audio: A Systematic Review

引用

IEEE Journal of Biomedical and Health Informatics 2025年

作者： Li, Yuxin Kumbale, Sinchana Chen, Yanru Surana, Tanmay Chng, Eng Siong Guan, Cuntai Nanyang Technological University College of Computing and Data Science Singapore Manipal Institute of Technology Department of Computer Science and Engineering Manipal India

Depression is a prevalent mental health disorder that presents significant challenges for timely diagnosis and intervention. Automated Depression Detection (ADD) systems using text and audio offer scalable mental health assessment solutions. This review systematically evaluates 65 studies published between 2018 and 2024, focusing on ADD methods that utilize machine learning models with multimodal data. We examine key methodologies, including data augmentation, multimodal fusion, and feature extraction, along with state-of-the-art ADD systems. The review emphasizes the need for culturally adaptable, high-quality datasets and interpretable models for clinical use. We also identify gaps in longitudinal data and real-world applications. Future research should focus on developing clinically integrated, cross-cultural ADD systems that are interpretable, scalable, and robust. The findings of this review contribute to the research field by providing a comprehensive overview of existing methodologies, identifying gaps in the current literature, and offering insights for future advancements in depression detection using speech and text analysis. © 2013 IEEE.

关键词： mHealth

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：