检索结果-内蒙古大学图书馆

Proceedings of the 36th International Conference on Neural Information Processing systems

作者： Biraj Dahal Alex Havrilla Minshuo Chen Tuo Zhao Wenjing Liao School of Mathematics Georgia Institute of Technology Electrical and Computer Engineering Princeton University Industrial and Systems Engineering Georgia Institute of Technology

ISBN: (纸本)9781713871088

Generative networks have experienced great empirical successes in distribution learning. Many existing experiments have demonstrated that generative networks can generate high-dimensional complex data from a low-dimensional easy-to- sample distribution. However, this phenomenon can not be justified by existing theories. The widely held manifold hypothesis speculates that real-world data sets, such as natural images and signals, exhibit low-dimensional geometric structures. In this paper, we take such low-dimensional data structures into consideration by assuming that data distributions are supported on a low-dimensional manifold. We prove statistical guarantees of generative networks under the Wasserstein-1 loss. We show that the Wasserstein-1 loss converges to zero at a fast rate depending on the intrinsic dimension instead of the ambient data dimension. Our theory leverages the low-dimensional geometric structures in data sets and justifies the practical power of generative networks. We require no smoothness assumptions on the data distribution which is desirable in practice.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dandelion Center Detection in Perennial Ryegrass with Heat Maps Using Convolutional Neural Networks

SSRN

引用

SSRN 2024年

作者： Babiker, Ibrahim Bentahar, Jamal Xie, Wen-Fang The Department of Mechanical Industrial and Aerospace Engineering Concordia University MontréalQC Canada Concordia Institute for Information Systems Engineering Concordia University MontréalQC Canada Electrical Engineering and Computer Science Department Khalifa University Abu Dhabi United Arab Emirates

industrial agricultural practices largely involve automated weed control processes and techniques. Carryover from this phenomenon can be seen in horticulture and landscaping. Within these fields, dandelion weeds (Taraxacum officinale) are a common pest and their detection is necessary for any type of removal. To address this detection issue, a cost-effective and intuitive labeling method is proposed for marking dandelion plant centers within perennial rye-grass. Furthermore, an expandable lightweight Convolutional Neural Network (CNN) is built on a base network to generate detection output maps at two resolutions. Multiple loss functions are formulated and their combinations are examined through ablation to assess and rank their performance. Different methods of computing standard metrics are also explored. By using these efficient methods, dandelion weed centers can consistently be located with relative robustness to noise and erroneous labels and with reasonable accuracy. The experimental results demonstrate that our methods outperform semantic segmentation models in terms of precision while avoiding the need for their intensive labeling costs. Our solution can be used in the landscaping industry and adapted to other fields with relative ease. The binary classification and object detection tasks of locating dandelion plants can be extended to multi-class problems with other plants. © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Routing Using Novel Optimization with Tabu Techniques for Wireless Sensor Network

引用

computer systems Science & engineering 2023年第5期45卷 1711-1726页

作者： Manar Ahmed Hamza Aisha Hassan Abdalla Hashim Dalia H.Elkamchouchi Nadhem Nemri Jaber S.Alzahrani Amira Sayed A.Aziz Mnahel Ahmed Ibrahim Abdelwahed Motwakel Department of Electrical and Computer Engineering International Islamic University MalaysiaKuala Lumpur53100Malaysia Department of Computer and Self Development Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Information Systems College of Science&Art at MahayilKing Khalid UniversitySaudi Arabia Department of Industrial Engineering College of Engineering at AlqunfudahUmm Al-Qura UniversitySaudi Arabia Department of Digital Media Faculty of Computers and Information TechnologyFuture University in EgyptNew Cairo11835Egypt College of Science and Humanities Studies in Alquwayiyah Shaqra UniversitySaudi Arabia

Wireless Sensor Network(WSN)consists of a group of limited energy source sensors that are installed in a particular region to collect data from the *** the energy-efficient data collection methods in largescale wireless sensor networks is considered to be a difficult area in the *** node clustering is a popular approach for ***,the sensor nodes are grouped to form clusters in a cluster-based WSN *** battery performance of the sensor nodes is likewise *** a result,the energy efficiency of WSNs is *** specific,the energy usage is influenced by the loads on the sensor node as well as it ranges from the Base Station(BS).Therefore,energy efficiency and load balancing are very essential in *** the proposed method,a novel Grey Wolf Improved Particle Swarm Optimization with Tabu Search Techniques(GW-IPSO-TS)was *** selection of Cluster Heads(CHs)and routing path of every CH from the base station is enhanced by the proposed *** provides the best routing path and increases the lifetime and energy efficiency of the ***-to-end delay and packet loss rate have also been *** proposed GW-IPSO-TS method enhances the evaluation of alive nodes,dead nodes,network survival index,convergence rate,and standard deviation of sensor *** to the existing algorithms,the proposed method outperforms better and improves the lifetime of the network.

关键词： Wireless sensor networks energy-efficient load balancing energy consumption network’s lifetime cluster heads grey wolf optimization tabu search particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Binary Codes for Correcting Asymmetric Adjacent Transpositions and Deletions

引用

IEEE Transactions on Communications 2024年

作者： Wang, Shuche Vu, Van Khu Tan, Vincent Y. F. National University Of Singapore Singapore Institute Of Operations Research And Analytics Singapore National University Of Singapore Singapore Department Of Industrial Systems Engineering And Management Singapore National University Of Singapore Singapore Department Of Mathematics Singapore National University Of Singapore Singapore Department Of Electrical And Computer Engineering Singapore

Codes in the Damerau-Levenshtein metric have received some attention by the research community recently owing to their applications in DNA-based data storage. In particular, Gabrys, Yaakobi, and Milenkovic (2017) designed a length-n code correcting a single deletion and s adjacent transpositions with at most (1+2s) log n bits of redundancy. In this work, we consider a new setting where both deletions and asymmetric adjacent transpositions may occur. For asymmetric transpositions, at most s+ 0-right shifts (i.e., 01 → 10) and at most s- 0-left shifts (i.e., 10 → 01) may occur. We present several constructions of the binary codes correcting these errors in various cases. In particular, we design a code correcting a single deletion, s+ right-shift, and s- left-shift errors with at most (1 + s) log(n + s + 1) + 1 bits of redundancy where s = s+ + s-. In addition, we investigate uniquely-decodable codes correcting t 0-deletions and s adjacent transpositions with at most (t + 2s) log n + o(log n) bits of redundancy. Then, we study the code for correcting t 0-deletions, s+ right-shift, and s- left-shift errors with list-decoding algorithms. Our main contribution here is the construction of a list-decodable code with list size O(ns) and with at most (max{t, s + 1}) log n + O(1) bits of redundancy, where s = s+ + s-. We construct non-systematic codes for correcting tb blocks of 0-deletions with l-limited magnitude and s adjacent transpositions with redundancy at most (2(tb+2s)+1) log(n+1)+O(1) bits and systematic codes with at most ((2(tb + 2s) + 1)(1 + 1/(log(tbl + 4))) log(N + 1) + O(log log N) redundant bits in an N-length codeword. © 1972-2012 IEEE.

关键词： Binary codes

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of Decision Tree, AdaBoost, Random Forest, Naïve Bayes, KNN, and Perceptron for Heart Disease Prediction

Comparative Study of Decision Tree, AdaBoost, Random Forest,...

引用

IEEE Southeastcon

作者： Mohammad Maydanchi Armin Ziaei Mina Basiri Alireza Norouzi Azad Shaheen Pouya Mehrbod Ziaei Fatemeh Haji Saman Sargolzaei Department of Industrial and Systems Engineering Auburn University AL USA Department of Engineering and Computer Science The University of Texas at Dallas TX USA Department of Industrial and Systems Engineering Tarbiat Modares University Tehran Iran Department of Engineering Science University of Tehran Tehran Iran Department of Electrical Engineering University of Science and Culture Tehran Iran Department of Computer Engineering Ferdowsi University of Mashhad Mashhad Iran Department of Engineering The University of Tennessee at Martin TN USA

Globally, cardiovascular diseases (CVD) are estimated to account for more than 32% of all deaths. Consequently, CVD has become a global health problem, and timely diagnosis is essential (WHO, 2021). Screening for risk factors accelerates the diagnosis and management of CVD, resulting in a more effective and rapid response, reducing the risk of death. This article compares six classification models, AdaBoost, Random Forest, Decision Tree, KNN, Naive Bayes, and Perceptron, to predict CVD symptoms. Based on CDC data collected from Kaggle, classification models were compared with the approach of examining effective factors to predict heart disease. Since the data set was imbalanced, the study performance was measured by AUC and F 1 -score in Class 1, which is the critical class in this dataset. AdaBoost is found to have the highest AUC and F 1 -score, respectively, of 0.828 and 0.37, while Decision Tree has the lowest AUC of 0.595 and F 1 -score of 0.25.

关键词： Heart Neural networks Forestry Predictive models Data models Decision trees Medical diagnosis

来源：评论

学校读者我要写书评

暂无评论

BEYOND ALGORITHMIC FAIRNESS: A GUIDE TO DEVELOP AND DEPLOY ETHICAL AI-ENABLED DECISION-SUPPORT TOOLS

arXiv

引用

arXiv 2024年

作者： Gonzalez, Rosemarie Santa Piansky, Ryan Bae, Sue M. Biddle, Justin Molzahn, Daniel H. Milton Stewart School of Industrial and Systems Engineering Georgia Institute of Technology United States School of Electrical and Computer Engineering Georgia Institute of Technology United States School of Public Policy Georgia Institute of Technology United States

The integration of artificial intelligence (AI) and optimization hold substantial promise for improving the efficiency, reliability, and resilience of engineered systems. Due to the networked nature of many engineered systems, ethically deploying methodologies at this intersection poses challenges that are distinct from other AI settings, thus motivating the development of ethical guidelines tailored to AI-enabled optimization. This paper highlights the need to go beyond fairness-driven algorithms to systematically address ethical decisions spanning the stages of modeling, data curation, results analysis, and implementation of optimization-based decision support tools. Accordingly, this paper identifies ethical considerations required when deploying algorithms at the intersection of AI and optimization via case studies in power systems as well as supply chain and logistics. Rather than providing a prescriptive set of rules, this paper aims to foster reflection and awareness among researchers and encourage consideration of ethical implications at every step of the decision-making process. Copyright © 2024, 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

Optimization and Optimizers for Adversarial Robustness

arXiv

引用

arXiv 2023年

作者： Liang, Hengyue Liang, Buyun Peng, Le Cui, Ying Mitchell, Tim Sun, Ju Department of Electrical & Computer Engineering University of Minnesota United States Department of Computer Science & Engineering University of Minnesota United States Department of Industrial & Systems Engineering University of Minnesota United States Department of Computer Science Queens College City University of New York United States

Empirical robustness evaluation (RE) of deep learning models against adversarial perturbations entails solving nontrivial constrained optimization problems. Existing numerical algorithms that are commonly used to solve them in practice predominantly rely on projected gradient, and mostly handle perturbations modeled by the l1, l2 and l∞ distances. In this paper, we introduce a novel algorithmic framework that blends a general-purpose constrained-optimization solver PyGRANSO with Constraint-Folding (PWCF), which can add more reliability and generality to the state-of-the-art RE packages, e.g., AutoAttack. Regarding reliability, PWCF provides solutions with stationarity measures and feasibility tests to assess the solution quality. For generality, PWCF can handle perturbation models that are typically inaccessible to the existing projected gradient methods;the main requirement is the distance metric to be almost everywhere differentiable. Taking advantage of PWCF and other existing numerical algorithms, we further explore the distinct patterns in the solutions found for solving these optimization problems using various combinations of losses, perturbation models, and optimization algorithms. We then discuss the implications of these patterns on the current robustness evaluation and adversarial training. Copyright © 2023, The Authors. All rights reserved.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Methods for Predicting Heart Disease: Neural Network, XGBoost, Gradient Boost, Logistic Regression, and SVM

A Comparison of Methods for Predicting Heart Disease: Neural...

引用

2025 IEEE SoutheastCon, SoutheastCon 2025

作者： Maydanchi, Mohammad Basiri, Mina Ziaei, Armin Ziaei, Mehrbod Izadi, Alireza Sargolzaei, Saman Auburn University Department of Industrial and Systems Engineering AL United States Tarbiat Modares University Industrial Engineering Department Tehran Iran The University of Texas Department of Engineering and Computer Science DallasTX United States University of Science and Culture Department of Electrical Engineering Tehran Iran University of Texas Department of Kinesiology San AntonioTX United States The University of Tennessee Department of Engineering MartinTN United States

ISBN: (纸本)9798331504847

Cardiovascular diseases (CVD) are a major cause of mortality, demanding immediate attention. Early screening for risk factors speeds up CVD identification and treatment, reducing mortality risk efficiently. This study compares the predictive abilities of XGBoost, Gradient Boost, Neural Network, Logistic Regression, and SVM in forecasting CVD symptoms using CDC data from Kaggle. Evaluating against a heart disease risk approach, we assess model performance with AUC and F1 - score for class 1 due to data imbalance. Additionally, we consider training run time as another metric for evaluation. XGBoost, Gradient Boost, and Logistic Regression excel in AUC and runtime, outperforming Neural Network and SVM significantly. The performances are quite similar when comparing the F1 - scores, but Logistic Regression, XGBoost, and Gradient Boost still maintain their superiority. The paper recommends using Logistic Regression and boosting-based decision tree models for future heart disease prediction investigations. © 2025 IEEE.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Selective Detection in Impulsive Low-Frequency Raman Imaging Using Shaped Probe Pulses

引用

Physical Review Applied 2023年第5期19卷 054075-054075页

作者： Siddarth Shivkumar Dekel Ranann Samuel Metais Sisira Suresh Nicolas Forget Randy Bartels Dan Oron Hervé Rigneault Aix Marseille Université CNRS Centrale Marseille Institut Fresnel Marseille France Department of Physics of Complex Systems The Weizmann Institute of Science Rehovot 76100 Israel Fastlite Antibes 06600 France Departments of Electrical and Computer Engineering and Biomedical Engineering Colorado State University Fort Collins USA

Impulsive stimulated Raman scattering (ISRS) using a single short femtosecond pump pulse to excite molecular vibrations offers an elegant pump-probe approach to perform vibrational imaging below 200cm−1. One shortcoming of ISRS is its inability to offer vibrational selectivity as all the vibrational bonds whose frequencies lie within the short pump-pulse bandwidth are excited. To date, several coherent control techniques have been explored to address this issue and selectively excite a specific molecular vibration by shaping the pump pulse. There has not been any systematic work that reports an analogous shaping of the probe pulse to implement preferential detection. In this work, we focus on vibrational imaging and report vibrational selective detection by shaping the probe pulse in time. We demonstrate numerically and experimentally two pulse-shaping strategies with one functioning as a vibrational notch filter and the other functioning as a vibrational low-pass filter. This enables fast (25μs/pixel) and selective hyperspectral imaging in the low-frequency regime (<200cm−1).

关键词： Analytical chemistry Molecular spectra Optical transient phenomena Biomolecules Time-resolved light scattering spectroscopy Ultrafast femtosecond pump probe

来源：评论

学校读者我要写书评

暂无评论

Pediatric Bone Age Prediction Using Deep Learning

Pediatric Bone Age Prediction Using Deep Learning

引用

International Conference on computer and Information Technology (ICCIT)

作者： Al Zadid Sultan Bin Habib Md. Ekramul Islam Md Asif Bin Syed Md Younus Ahamed Tanpia Tasnim Lane Department of Computer Science and Electrical Engineering West Virginia University Morgantown USA Department of Computer Science & Engineering Stamford University Bangladesh Dhaka Bangladesh Department of Industrial and Management Systems Engineering West Virginia University Morgantown USA Department of Computer Science and Engineering Green University of Bangladesh Dhaka Bangladesh

Pediatric bone age prediction is a crucial task in clinical practice that can help diagnose endocrine disorders and provide insight into a child’s growth and development. However, conventional bone age prediction methods are often labor-intensive and require specialized radiological expertise. This paper presents a Deep Learning (DL)-based approach to pediatric bone age prediction using EfficientNet with Additive Attention, a state-of-the-art neural network architecture for image classification and regression tasks. The method utilizes over 12,000 X-ray images from the RSNA bone age dataset. It involves image preprocessing, transforming them into three-channel images, and training a Convolutional Neural Network (CNN) to automatically learn the features of hand bone images. This approach provides a more effective and accurate solution for predicting bone age, which is critical in diagnosing pediatric endocrine diseases. This work uses two variations of the EfficientNet model (B0 and B4), where EfficientNetB4 is also finetuned with the Additive Attention mechanism. These three models predict the age for the original age, and their comparison is shown in curves. The predicted ages depict that in most cases, EfficientNetB4 and EfficientNetB4 with Additive Attention (EN-AA) successfully predicted the bone ages more accurately regarding the original age, and their performance was better than the EfficientNetB0. Specific performance metrics are provided to underscore this improvement. Learning curves for training and validation loss confirm effective learning without overfitting or underfitting, further validating our approach’s efficacy in pediatric endocrine disease diagnosis.

关键词：

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：