检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Hongjian Dandapanthula, Sanjit Ramdas, Aaditya Department of Statistics and Data Science Carnegie Mellon University United States Machine Learning Department Carnegie Mellon University United States

The recent e-Benjamini-Hochberg (e-BH) procedure for multiple hypothesis testing is known to control the false discovery rate (FDR) under arbitrary dependence between the input e-values. This paper points out an important subtlety when applying the e-BH procedure with e-processes, which are sequential generalizations of e-values (where the data are observed sequentially). Since adaptively stopped e-processes are e-values, the e-BH procedure can be repeatedly applied at every time step, and one can continuously monitor the e-processes and the rejection sets obtained. One would hope that the "stopped e-BH procedure" (se-BH) has an FDR guarantee for the rejection set obtained at any stopping time. However, while this is true if the data in different streams are independent, it is not true in full generality, because each stopped e-process is an e-value only for stopping times in its own local filtration, but the se-BH procedure employs a stopping time with respect to a global filtration. This can cause information to leak across time, allowing one stream to know its future by knowing past data of another stream. This paper formulates a simple causal condition under which local e-processes are also global e-processes and thus the se-BH procedure does indeed control the FDR. The condition excludes unobserved confounding from the past and is met under most reasonable scenarios including genomics. Copyright © 2025, The Authors. All rights reserved.

关键词： Filtration

来源：评论

学校读者我要写书评

暂无评论

Automatic Human Life Rescue Analysis Model with Incorporation of Deep learning Algorithm 5

Automatic Human Life Rescue Analysis Model with Incorporatio...

引用

5th International Conference on Sustainable Communication Networks and Application, ICSCNA 2024

作者： Kalaiselvi, V.K.G. Prithikhaa, S. Divya Darshini, J. Reddy, S Bharath Anandkumar, B. Hariharan, Shanmugasundaram Sri Sairam Engineering College Department of Information Technology Tamilnadu India KG Reddy College of Engineering and Technology Department of CSE Telangana India Vardhaman College of Engineering Department of Artificial Intelligence and Machine Learning India Vardhaman College of Engineering Department of Artificial Intelligence and Data Science Hyderabad India

ISBN: (纸本)9798331530013

In today's fast-paced world, every passing moment can mean the difference between life and death. To address this critical issue, we have developed an innovative application designed to provide rapid emergency medical response. Ambulance services play a pivotal role in emergency healthcare by offering vital transportation and care. However, limited access to ambulance contact information often hampers patients' ability to secure timely assistance during emergencies. With the widespread adoption of smartphones and mobile applications, the proposed solution integrates React Native with a Convolutional Neural Network (CNN) to create an advanced mobile app for law enforcement, capable of real-time car accident detection. Utilizing datasets from Kaggle, images are resized to optimize CNN training. Through transfer learning, the CNN architecture effectively extracts spatial features, enabling accurate accident prediction, validated by robust evaluation metrics such as precision and recall. The React Native app facilitates swift notifications to law enforcement, utilizing cutting-edge deep learning technology to enhance road safety and improve emergency response efficiency. This project aims to streamline emergency services, minimizing response and transport times, thereby saving lives and fostering safer communities. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Statistical guarantees for local spectral clustering on random neighborhood graphs

引用

Journal of machine learning Research 2021年第1期22卷 1-71页

作者： Green, Alden Balakrishnan, Sivaraman Tibshirani, Ryan J. Department of Statistics and Data Science Carnegie Mellon University PittsburghPA15213 United States Department of Statistics and Data Science Machine Learning Department Carnegie Mellon University PittsburghPA15213 United States

We study the Personalized PageRank (PPR) algorithm, a local spectral method for clustering, which extracts clusters using locally-biased random walks around a given seed node. In contrast to previous work, we adopt a classical statistical learning setup, where we obtain samples from an unknown nonparametric distribution, and aim to identify sufficiently salient clusters. We introduce a trio of population-level functionals—the normalized cut, conductance, and local spread, analogous to graph-based functionals of the same name—and prove that PPR, run on a neighborhood graph, recovers clusters with small population normalized cut and large conductance and local spread. We apply our general theory to establish that PPR identifies connected regions of high density (density clusters) that satisfy a set of natural geometric conditions. We also show a converse result, that PPR can fail to recover geometrically poorly-conditioned density clusters, even asymptotically. Finally, we provide empirical support for our theory. ©2021 Alden Green, Sivaraman Balakrishnan, and Ryan J. Tibshirani.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Using Generative Models to Improve Fire Detection Efficiency

Using Generative Models to Improve Fire Detection Efficiency

引用

Information Technology and Nanotechnology (ITNT), International Conference on

作者： Nikita Andriyanov Alexandr Kim Xenin Fao Data Analysis and Machine Learning Department Financial University under the Government of the Russian Federation Moscow Russia Data Science Department Huazhong University of Science and Technology Huazhong China

ISBN: (数字)9798350372939

ISBN: (纸本)9798350372946

The paper discusses generative artificial intelligence technologies used to improve the efficiency of fire detection in satellite images. Different detector architectures are proposed and compared in terms of accuracy. It is found that the use of generative models to expand the training sample is on average 1–2% more effective than the use of standard augmentations. The combined method of expanding the training image base can improve the accuracy by 4-5%. In addition, the right approach in generative prompts can also improve the accuracy metrics of the models.

关键词： Training Measurement YOLO Accuracy Image databases Detectors Satellite images

来源：评论

学校读者我要写书评

暂无评论

An Improved Finite-time Analysis of Temporal Difference learning with Deep Neural Networks 41

An Improved Finite-time Analysis of Temporal Difference Lear...

引用

41st International Conference on machine learning, ICML 2024

作者： Ke, Zhifa Wen, Zaiwen Zhang, Junyu Center for Data Science Peking University China Beijing International Center for Mathematical Research Center for Machine Learning Research Changsha Institute for Computing and Digital Economy Beijing China Department of Industrial Systems Engineering and Management National University of Singapore Singapore

Temporal difference (TD) learning algorithms with neural network function parameterization have well-established empirical success in many practical large-scale reinforcement learning tasks. However, theoretical understanding of these algorithms remains challenging due to the nonlinearity of the action-value approximation. In this paper, we develop an improved non-asymptotic analysis of the neural TD method with a general L-layer neural network. New proof techniques are developed and an improved new Õ(ϵ-1) sample complexity is derived. To our best knowledge, this is the first finite-time analysis of neural TD that achieves an Õ(ϵ-1) complexity under the Markovian sampling, as opposed to the best known Õ(ϵ-2) complexity in the existing literature. Copyright 2024 by the author(s)

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Active multiple testing with proxy p-values and e-values

arXiv

引用

arXiv 2025年

作者： Xu, Ziyu Wang, Catherine Wasserman, Larry Roeder, Kathryn Ramdas, Aaditya Department of Statistics and Data Science United States Machine Learning Department Germany Computational Biology Department Carnegie Mellon University United States

Researchers often lack the resources to test every hypothesis of interest directly or compute test statistics comprehensively, but often possess auxiliary data from which we can compute an estimate of the experimental outcome. We introduce a novel approach for selecting which hypotheses to query a statistic (i.e., run an experiment, perform expensive computation, etc.) in a hypothesis testing setup by leveraging estimates (e.g., from experts, machine learning models, previous experiments, etc.) to compute proxy statistics. Our framework allows a scientist to propose a a proxy statistic, and then query the true statistic with some probability based on the value of the proxy. We make no assumptions about how the proxy is derived and it can be arbitrarily dependent with the true statistic. If the true statistic is not queried, the proxy is used in its place. We characterize "active" methods that produce valid p-values and e-values in this setting and utilize this framework in the multiple testing setting to create procedures with false discovery rate (FDR) control. Through simulations and real data analysis of causal effects in scCRISPR screen experiments, we empirically demonstrate that our proxy framework has both high power and low resource usage when our proxies are accurate estimates of the respective true statistics. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Model and Feature Diversity for Bayesian Neural Networks in Mutual learning 37

Model and Feature Diversity for Bayesian Neural Networks in ...

引用

37th Conference on Neural Information Processing Systems, NeurIPS 2023

作者： Pham, Cuong Nguyen, Cuong C. Le, Trung Phung, Dinh Carneiro, Gustavo Do, Thanh-Toan Department of Data Science and AI Monash University Australia Australian Institute for Machine Learning University of Adelaide Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom VinAI Viet Nam

ISBN: (纸本)9781713899921

Bayesian Neural Networks (BNNs) offer probability distributions for model parameters, enabling uncertainty quantification in predictions. However, they often underperform compared to deterministic neural networks. Utilizing mutual learning can effectively enhance the performance of peer BNNs. In this paper, we propose a novel approach to improve BNNs performance through deep mutual learning. The proposed approaches aim to increase diversity in both network parameter distributions and feature distributions, promoting peer networks to acquire distinct features that capture different characteristics of the input, which enhances the effectiveness of mutual learning. Experimental results demonstrate significant improvements in the classification accuracy, negative log-likelihood, and expected calibration error when compared to traditional mutual learning for BNNs. © 2023 Neural information processing systems foundation. All rights reserved.

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review of Supervised learning Algorithms in Healthcare Applications 2

A Comprehensive Review of Supervised Learning Algorithms in ...

引用

2nd DMIHER International Conference on Artificial Intelligence in Healthcare, Education and Industry, IDICAIEI 2024

作者： Barhate, Aditya Tale, Abhay Jikar, Nayan Verma, Prateek Kumar, Praveen Yesankar, Prajyot Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artificial Intelligence and Machine Learning Maharashtra Wardha India Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artificial Intelligence and Data Science Maharashtra Wardha India

ISBN: (纸本)9798331528713

Supervised learning has revolutionized the concept of personalization in treatment with the development of Precision Medicine. This review aims to provide a systematic analysis of the utilization of supervised learning in healthcare and its functions in diagnostics and prognoses, recommendations for treatment, and patient surveillance. Classification algorithms such as Logistic Regression, Support Vector machines, and Decision Trees. Regression techniques such as Linear Regression, Polynomial Regression, and Neural Networks. Ensemble methods such as Bagging, Boosting, and Stacking. Deep learning techniques such as Convolutional Neural Networks, Recurrent Neural Networks, and Long Short-Term Memory are discussed for their efficiency and applicability in various healthcare applications. There are strengths highlighted by the review including enhanced diagnostic capabilities and patient-targeted therapy alongside weaknesses such as quality of data, interpretability as well as compatibility with existing healthcare systems. This review gives a more generalized approach to how supervised learning algorithms help to enhance healthcare results and identifies the likely topic that requires more research in the future. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

machine learning-Enhanced Self-Management for Energy-Effective and Secure Statistics Assortment in Unattended WSNs

引用

SN Computer science 2025年第2期6卷 1-10页

作者： Biswas, Preesat Mishra, Aishwarya Dixit, Rinku Sharma Dwivedi, Abhishek Choudhary, Shailee Lohmor Tiwari, Shweta Chakravarthi, M. Ayyappa Electronics & Telecommunications Government Engineering College C. G Jagdalpur India Department of Computer Science & Engineering IES College of Technology M.P Bhopal India Department of Artificial Intelligence and Machine Learning New Delhi Institute of Management Delhi India Department of Data Science and Engineering School of AIML IoT & IS CCE DS and Computer Applications Manipal University Jaipur Jaipur Rajasthan India Department of Computer Science and Engineering Gyan Ganga Institute of Technology and Sciences M.P Jabalpur India Department of CSE-Data Science KKR and KSR Institute of Technology and Sciences Guntur India

Unattended Wireless Sensor Networks (UWSNs) operate without human supervision and have limited resources. In a conventional WSN with a fixed sink for data collection, nodes one hop away from the sink consume more energy, causing a hotspot issue. This leads to network fragmentation and reduced sensing efficiency. To address this, a mobile sink is proposed. This study details the development and implementation of Secure MRL, focusing on legitimate message communication to the mobile sink via convex nodes. By using these convex nodes as data collection points, the mobile sink avoids traversing the entire network. We employ an enhanced system to generate keys and authenticate messages, assessing its resilience against various attacks, including sleep deprivation, snooze, network substitution, and modification. Findings indicate that Secure MRL maintains consistent performance across different node counts and significantly reduces energy consumption by 84%, improving the speed of attack detection. This improvement is achieved by monitoring the digital signature and residual energy of nodes. The proposed SVM-based network training establishes a residual energy cut-off, while convex nodes mitigate synchronization issues among other sensor nodes in Secure MRL. Additionally, convex nodes manage the availability checks for the mobile sink. The proposed approach demonstrates a 92% improvement in packet delivery ratio (PDR) and an 89% reduction in end-to-end delay, highlighting its efficacy in enhancing network performance. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Duty-cycled UWSN MRL Sec-MRL SEC-TMP SVM UWSN

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Product Quality Evaluation using Hybrid LSTM-FCNN Model and GWO Algorithm for Social Media Sentiment Analysis 8

An Enhanced Product Quality Evaluation using Hybrid LSTM-FCN...

引用

8th International Conference on Electronics, Communication and Aerospace Technology, ICECA 2024

作者： Raj Kumar, V.S. Kumaresan, T. Kanna, P. Rajesh Jagadeesan, S. Showmiya, P. Nithin, P. Bannari Amman Institute of Technology Department of Artificial Intelligence and Data Science Erode India Bannari Amman Institute of Technology Department of Computer Science and Engineering Erode India Nandha Engineering College Department of Computer Science and Engineering Erode India Nehru Institute of Technology Department of Computer Science - Cyber Security Coimbatore India Bannari Amman Institute of Technology Department of Artificial Intelligence and Machine Learning Erode India

ISBN: (纸本)9798350367904

Sentiment analysis is still in its developing era, and sometimes, struggles are faced in evaluating user sentiment due to the use of traditional models to analyze the relationship structures in social media interactions. To tackle the issue, we proposed an innovative approach Hybrid LSTM neural network and a fully connected neural network (FCNN) integrated with the Grey Wolf Optimization (GWO) algorithm. The objective of this proposed method is to improve sentiment analysis by employing the LSTM to capture the sequence of data and improve the ability of FCNNs to improve feature extraction and the analysis process. The proposed system process initiated with the data preprocessing of input Twitter sentiment dataset transformed into a suitable input to the Hybrid LSTM-FCNN neural network. Then, temporal dependency of the data captured by LSTM and FCNN is used to extract the features. The GWO method mimics the grey wolf hunting behaviour, for optimizing the proposed Hybrid LSTM-FCNN model's parameters to enhance the classification accuracy. The system models start with the initialization of the model, parameters, and training the models using the dataset to reduce the cross-entropy loss, estimating the testing dataset from unseen data and fine-tuning using GWO as necessary. Results demonstrate significant improvements in sentiment categorization compared to traditional methods. The Hybrid LSTM-FCNN model, enhanced with GWO, effectively captures nuanced sentiment patterns, leading to more accurate classification. Additionally, GWO's computational efficiency contributes to scalability. In conclusion, our proposed approach offers a promising solution for enhancing product quality evaluation through advanced neural network architectures and optimization algorithms applied to social media sentiment analysis. © 2024 IEEE.

关键词： Tweets

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：