检索结果-内蒙古大学图书馆

Innovations in High Speed Communication and Signal Processing (IHCSP), International Conference on

作者： Divya Nimma Thamara Selvi M D S.K Sunori Punit Pathak Sabareesh R Prema S PhD in Computational Science University of Southern Mississippi Data Analyst in UMMC USA Department of Verbal and Life Skills Saveetha School of Engineering SIMATS Engineering Thandalam Chennai India Department of Electronics and Communication Engineering Graphic Era Hill University Bhimtal Department of School of Liberal Arts and Human Sciences Auro University Surat India Skill Development & Employability Programs Fortuna Business Solutions (R&D) Coimbatore India Department of English Panimalar Engineering College Chennai India

ISBN: (数字)9798350368949

ISBN: (纸本)9798350368956

Predicting student achievement in English language learning has traditionally relied on standardized tests and trainer opinions. While these methods provide precious insights, they often fail to capture nuanced components of language talent and character gaining knowledge of desires. Existing techniques commonly contain static checks that don't adapt to the evolving nature of scholar overall performance or provide personalized feedback. This study proposes a singular approach utilizing GPT-3 (Generative Pre-skilled Transformer) models to enhance the prediction of scholar fulfillment in English language learning. GPT-3 models, regarded for their superior natural language processing skills, provide a dynamic and personalized evaluation of scholar texts. Unlike traditional methods, GPT-three can compare complex language capabilities, including grammar, vocabulary utilization, and coherence, and generate tailored remarks that addresses precise areas of development. The methodology entails amassing a various dataset of scholar-written texts and engagement metrics. These data are preprocessed and used to high-quality-song a GPT-three version, allowing it to assess language talent and predict destiny performance. The satisfactory-tuned version analyzes student submissions, generates personalized feedback, and predicts fulfillment primarily based on found styles and contextual know-how. Results exhibit that the GPT-3 based approach substantially improves the accuracy of fulfillment predictions and offers greater actionable feedback as compared to conventional methods. The version’s potential to generate real-time, contextually relevant remarks complements the gaining knowledge of revel in by using addressing man or woman wishes and adapting to student progress.

关键词： Measurement Adaptation models Vocabulary Technological innovation Signal processing Predictive models Transformers Natural language processing Real-time systems Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

Discovering High-Entropy Oxides with a Machine-Learning Interatomic Potential

arXiv

引用

arXiv 2024年

作者： Sivak, Jacob T. Almishal, Saeed S.I. Caucci, Mary K. Tan, Yueze Srikanth, Dhiya Furst, Matthew Chen, Long-Qing Rost, Christina M. Maria, Jon-Paul Sinnott, Susan B. Department of Chemistry The Pennsylvania State University University ParkPA16802 United States Department of Materials Science and Engineering The Pennsylvania State University University ParkPA16802 United States Department of Mathematics The Pennsylvania State University University ParkPA16802 United States Department of Engineering Science and Mechanics The Pennsylvania State University University ParkPA16802 United States Department of Materials Science and Engineering Virginia Polytechnic Institute State University BlacksburgVA24061 United States Institute for Computational and Data Sciences The Pennsylvania State University University ParkPA16802 United States

High-entropy materials shift the traditional materials discovery paradigm to one that leverages disorder, enabling access to unique chemistries unreachable through enthalpy alone. We present a self-consistent approach integrating computation and experiment to understand and explore single-phase rock salt high-entropy oxides. By leveraging a machine-learning interatomic potential, we rapidly and accurately map high-entropy composition space using our two descriptors: bond length distribution and mixing enthalpy. The single-phase stabilities for all experimentally stabilized rock salt compositions are correctly resolved, with dozens more compositions awaiting discovery. © 2024, CC BY.

关键词： Enthalpy

来源：评论

学校读者我要写书评

暂无评论

Untangling individual cation roles in rock salt high-entropy oxides

arXiv

引用

arXiv 2024年

作者： Almishal, Saeed S.I. Sivak, Jacob T. Kotsonis, George N. Tan, Yueze Furst, Matthew Srikanth, Dhiya Crespi, Vincent H. Gopalan, Venkatraman Heron, John T. Chen, Long-Qing Rost, Christina M. Sinnott, Susan B. Maria, Jon-Paul Department of Materials Science and Engineering The Pennsylvania State University University ParkPA16802 United States Department of Chemistry The Pennsylvania State University University ParkPA16802 United States Department of Physics The Pennsylvania State University University ParkPA16802 United States Department of Materials Science and Engineering University of Michigan Ann ArborMI48109 United States Department of Materials Science and Engineering Virginia Polytechnic Institute and State University BlacksburgVA24061 United States Institute for Computational and Data Sciences The Pennsylvania State University University ParkPA16802 United States

We unravel the distinct roles each cation plays in phase evolution, stability, and properties within Mg1/5Co1/5Ni1/5Cu1/5Zn1/5O high-entropy oxide (HEO) by integrating experimental findings, thermodynamic analyses, and first-principles predictions. Our approach is through sequentially removing one cation at a time from the five-component high-entropy oxide to create five four-component derivatives. Bulk synthesis experiments indicate that Mg, Ni, and Co act as rock salt phase stabilizers whereas only Mg and Ni enthalpically enhance single-phase rock salt stability in thin film growth;synthesis conditions dictate whether Co is a rock salt phase stabilizer or destabilizer. By examining the competing phases and oxidation state preferences using pseudo-binary phase diagrams and first-principles calculations, we resolve the stability differences between bulk and thin film for all compositions. We systematically explore HEO macroscopic property sensitivity to cation selection employing both predicted and measured optical spectra. This study establishes a framework for understanding high-entropy oxide synthesizability and properties on a per-cation basis that is broadly applicable to tailoring functional property design in other high-entropy materials. © 2024, CC BY-NC-SA.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Z-Number-Valued Rule-Based Classification System

SSRN

引用

SSRN 2022年

作者： Li, Yangxue Herrera-Viedma, Enrique Pérez, Ignacio Javier Barragán-Guzmán, Mónica Morente-Molinera, Juan Antonio Andalusian Research Institute in Data Science and Computational Intelligence Dept. of Computer Science and AI University of Granada Granada18071 Spain Department of Electrical and Computer Engineering Faculty of Engineering King Abdulaziz University Jeddah21589 Saudi Arabia

The fuzzy rule-based classification system (FRBCS) is a popular tool in classification problems due to its interpretability and comprehensibility. As an extension of fuzzy numbers, the concept of Z-number is a more appropriate formal structure to describe uncertain and partially reliable information. A Z-number is an ordered pair of fuzzy numbers, where the second fuzzy number describes the reliability of the first one. As a result of its representation capability, it can receive better classification results. However, there is still a gap in the application of Z-numbers to classification problems due to its high computation complexity. To take advantage of the Z-number, we design a simple way to make Z-numbers apply to classification problems. Using the second fuzzy number to adjust the first fuzzy number to fit the unevenly distributed data. Then we create a Z-number-valued if-then rule by extending the fuzzy if-then rule. In addition, a Z-number-valued rule-based classification system (ZRBCS) is developed, including two main processes: rules generation and new pattern classification. Compared to the classic fuzzy rule-based system, the proposed system can cover more information that can improve classification results. We evaluated the proposed method on 20 datasets to compare its learning and predictive abilities against the classic fuzzy rule-based system. The results demonstrate the Z-number-valued rule-based classification system is superior to the fuzzy rule-based classification system. © 2022, The Authors. All rights reserved.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Advanced Fault Detection in Photovoltaic Panels Using Enhanced U-Net Architectures

SSRN

引用

SSRN 2025年

作者： Awedat, Khalfalla Comert, Gurcan Ayad, Mustafa Mrebit, Abdulmajid Computer Information Technology Department SUNY Morrisville College 80 Eaton St MorrisvilleNY13408 United States Computational Data Science and Engineering Department North Carolina A&T State University 1601 E. Market Street GreensboroNC27411 United States Electrical and Computer Engineering Department SUNY Oswego 430 Shineman Center OswegoNY13126 United States School of Art and Sciences Computer Science Physics & Engineering Department Benedict College 1600 Harden Street ColumbiaSC29204 United States

Fault detection in photovoltaic (PV) panels using thermal images remains a significant challenge due to the com-plexity of thermal patterns, environmental noise, and the subtle nature of anomalies. This paper introduces anadvanced deep learning framework that enhances the U-Net architecture by integrating Residual Blocks, AtrousSpatial Pyramid Pooling (ASPP), and Attention Mechanisms. These enhancements collectively improve featureextraction, contextual understanding, and fault localization, addressing the limitations of traditional segmentationapproaches and reducing false positives. Extensive experiments demonstrate that the proposed method significantlyoutperforms all benchmarked algorithms across key segmentation metrics, including standard U-Net, U-Net withASPP, and DeepLabV3+. Compared to standard U-Net, the proposed model achieves more than a 29% increasein F1-score and a 62% improvement in Intersection over Union (IoU) while reducing segmentation loss by 71%.Its ability to accurately detect faults under challenging conditions establishes the framework as a state-of-the-artsolution for real-time PV monitoring. These results demonstrate the effectiveness of the proposed approach in ad-dressing the challenges of PV fault detection, offering a practical and reliable solution for ensuring the operationalperformance of renewable energy systems. © 2025, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Crop Yield Prediction Using Spatio Temporal CNN and Multimodal Remote Sensing

Crop Yield Prediction Using Spatio Temporal CNN and Multimod...

引用

International Conference on Edge Computing and Applications (ICECAA)

作者： AR. Arunarani S. Selvanayaki Mohammed Saleh Al Ansari Md. Abul Ala Walid Nagamalleswari Devireddy Murugesan Mohana Keerthi Department of Computational Intelligence School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulatur Campus Chennai Tamilnadu India Artificial Intelligence and Data Science Saveetha Engineering College Chennai Tamilnadu India Department of Chemical Engineering College of Engineering University of Bahrain Bahrain Department of CSE Khulna University of Engineering & Technology (KUET) Khulna Bangladesh Department of Freshman Mohan Babu University Tirupathi Andhra Pradesh India School of Agriculture SR University Warangal Telangana India

Crop yield forecasting has been thoroughly studied recently and has become more important in preserving the safety of food. The majority of studies have concentrated on obtaining fixed spatial data from remote sensing images. Crop development, however, is a very complicated attribute influenced by a variety of factors. This work is intended to concurrently use spatial and temporal information from multimodal remotely sensed images in order to fully investigate these diverse characteristics. With the goal to make use of their supportive nature, Spatio temporal Convolutional Neural Networks (STCNN), is proposed as a revolutionary deep learning architecture for agricultural production prediction. To recognize the combined spatial-temporal representation, the STCNN specifically combines a spatial training segment and a temporal dependent segment into the convolutional network. The innovative spatial training segment begins by extracting enough spatial information from the multimodal spatial images. Then, in order to determine the temporal connection from the lengthy time-series images, the temporal dependent segment is concatenated on top of the spatial training segment. Both common machine learning techniques and cutting-edge deep learning techniques are compared to the outcomes of the proposed method. The experimental findings show that the STCNN can offer superior prediction performance of 0.64 RMSE and 12.03 MAPE values compared to the rival methods with minimized errors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Collaborative Learning of Discrete Distributions under Heterogeneity and Communication Constraints

arXiv

引用

arXiv 2022年

作者： Huang, Xinmeng Lee, Donghwan Dobriban, Edgar Hassani, Hamed Graduate Group in Applied Mathematics and Computational Science Univ. of Pennsylvania United States Department of Statistics and Data Science Univ. of Pennsylvania United States Department of Electrical and Systems Engineering Univ. of Pennsylvania United States

In modern machine learning, users often have to collaborate to learn the distribution of the data. Communication can be a significant bottleneck. Prior work has studied homogeneous users-i.e., whose data follow the same discrete distribution-and has provided optimal communication-efficient methods for estimating that distribution. However, these methods rely heavily on homogeneity, and are less applicable in the common case when users' discrete distributions are heterogeneous. Here we consider a natural and tractable model of heterogeneity, where users' discrete distributions only vary sparsely, on a small number of entries. We propose a novel two-stage method named SHIFT: First, the users collaborate by communicating with the server to learn a central distribution;relying on methods from robust statistics. Then, the learned central distribution is fine-tuned to estimate their respective individual distribution. We show that SHIFT is minimax optimal in our model of heterogeneity and under communication constraints. Further, we provide experimental results using both synthetic data and n-gram frequency estimation in the text domain, which corroborate its efficiency. Copyright © 2022, The Authors. All rights reserved.

关键词： Frequency estimation

来源：评论

学校读者我要写书评

暂无评论

Mining data impressions from deep models as substitute for the unavailable training data

arXiv

引用

arXiv 2021年

作者： Nayak, Gaurav Kumar Mopuri, Konda Reddy Jain, Saksham Chakraborty, Anirban Department of Computational and Data Sciences Indian Institute of Science Bangalore India Department of Computer Science and Engineering Indian Institute of Technology Tirupati

Pretrained deep models hold their learnt knowledge in the form of model parameters. These parameters act as "memory" for the trained models and help them generalize well on unseen data. However, in absence of training data, the utility of a trained model is merely limited to either inference or better initialization towards a target task. In this paper, we go further and extract synthetic data by leveraging the learnt model parameters. We dub them data Impressions, which act as proxy to the training data and can be used to realize a variety of tasks. These are useful in scenarios where only the pretrained models are available and the training data is not shared (e.g., due to privacy or sensitivity concerns). We show the applicability of data impressions in solving several computer vision tasks such as unsupervised domain adaptation, continual learning as well as knowledge distillation. We also study the adversarial robustness of lightweight models trained via knowledge distillation using these data impressions. Further, we demonstrate the efficacy of data impressions in generating data-free Universal Adversarial Perturbations (UAPs) with better fooling rates. Extensive experiments performed on benchmark datasets demonstrate competitive performance achieved using data impressions in absence of original training data. Copyright © 2021, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Utilizing Artificial Intelligence for Enhancing Performance and Preventing Injuries in Sports Analytics

Utilizing Artificial Intelligence for Enhancing Performance ...

引用

Intelligent Computing and Sustainable Innovations in Technology (IC-SIT), International Conference on

作者： Tushar Dhar Shukla Divya Nimma Kiran Sree Pokkuluri Syed Najmusaqib K.K. Sivakumar B Kiran Bala University Institute of Physical Education & Sports Chandigarh University Mohali India Computational Science University of Southern Mississippi Data Analyst in UMMC USA Department of CSE Shri Vishnu Engineering College for Women Bhimavaram India Department of Physical Education Central University of Kashmir India Mohan Babu University Tirupati AP India Department of AI & DS K.Ramakrishnan College of Engineering Trichy India

ISBN: (数字)9798350369175

ISBN: (纸本)9798350369182

The sports sector could see a revolution in player performance enhancement and injury prevention, especially with recent developments in artificial intelligence (AI). In order to identify areas for improvement and potential injury risks, this proposed work aims to utilize the potential of AI techniques, such as XGBoost, to evaluate extensive player data, including movement patterns, biomechanics, and physical condition. This strategy is unusual because it combines real-time feedback systems with AI-powered predictive modeling to offer athletes and coaches individualized training advice and early warnings about potential injury risks. Cleaning and normalization were done during the preparation phase to make sure the data was suitable for analysis. Utilizing XGBoost for feature extraction allowed for the identification of critical factors affecting efficiency and injury risk. In sports analytics, the suggested XGBoost-AI system was compared to other AI-based techniques for improving performance and avoiding injuries. The study’s findings show that the XGBoost classifier significantly improves performance accuracy, which may reach 96%. This is in contrast to conventional approaches that depend on subjective evaluations and human data analysis, which can result in errors and inefficiencies. Python is used to implement the suggested work. AI-driven sports analytics platform can improve game strategy, fan engagement, and data-driven decision-making for sports organizations, enhancing competitiveness and sustainability in the sports industry.

关键词： Training Technological innovation Prevention and mitigation Organizations Predictive models Real-time systems Artificial intelligence Sustainable development Injuries Sports

来源：评论

学校读者我要写书评

暂无评论

T-Cal: An optimal test for the calibration of predictive models

arXiv

引用

arXiv 2022年

作者： Lee, Donghwan Huang, Xinmeng Hassani, Hamed Dobriban, Edgar Graduate Group in Applied Mathematics and Computational Science Univ. of Pennsylvania United States Department of Electrical and Systems Engineering Univ. of Pennsylvania United States Department of Statistics and Data Science Univ. of Pennsylvania United States

The prediction accuracy of machine learning methods is steadily increasing, but the calibration of their uncertainty predictions poses a significant challenge. Numerous works focus on obtaining well-calibrated predictive models, but less is known about reliably assessing model calibration. This limits our ability to know when algorithms for improving calibration have a real effect, and when their improvements are merely artifacts due to random noise in finite datasets. In this work, we consider detecting mis-calibration of predictive models using a finite validation dataset as a hypothesis testing problem. The null hypothesis is that the predictive model is calibrated, while the alternative hypothesis is that the deviation from calibration is sufficiently large. We find that detecting mis-calibration is only possible when the conditional probabilities of the classes are sufficiently smooth functions of the predictions. When the conditional class probabilities are Hölder continuous, we propose T-Cal, a minimax optimal test for calibration based on a debiased plug-in estimator of the 2-Expected Calibration Error (ECE). We further propose adaptive T-Cal, a version that is adaptive to unknown smoothness. We verify our theoretical findings with a broad range of experiments, including with several popular deep neural net architectures and several standard post-hoc calibration methods. T-Cal is a practical general-purpose tool, which-combined with classical tests for discrete-valued predictors-can be used to test the calibration of virtually any probabilistic classification method. T-Cal is available at https://***/dh7401/T-Cal. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

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：