检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Leng, Jiaqi Ju, Yakun Duan, Yuanxu Zhang, Jiangnan Lv, Qingxuan Wu, Zuxuan Fan, Hao Haide College Ocean University of China China Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China School of Computing and Mathematic Sciences University of Leicester United Kingdom Department of Computer Engineering Qingdao City University China College of Computer Science and Technology Ocean University of China China

ISBN: (纸本)157735897X

Surface-from-gradients (SfG) aims to recover a three-dimensional (3D) surface from its gradients. Traditional methods encounter significant challenges in achieving high accuracy and handling high-resolution inputs, particularly facing the complex nature of discontinuities and the inefficiencies associated with large-scale linear solvers. Although recent advances in deep learning, such as photometric stereo, have enhanced normal estimation accuracy, they do not fully address the intricacies of gradient-based surface reconstruction. To overcome these limitations, we propose a Fourier neural operator-based Numerical Integration Network (FNIN) within a two-stage optimization framework. In the first stage, our approach employs an iterative architecture for numerical integration, harnessing an advanced Fourier neural operator to approximate the solution operator in Fourier space. Additionally, a self-learning attention mechanism is incorporated to effectively detect and handle discontinuities. In the second stage, we refine the surface reconstruction by formulating a weighted least squares problem, addressing the identified discontinuities rationally. Extensive experiments demonstrate that our method achieves significant improvements in both accuracy and efficiency compared to current state-of-the-art solvers. This is particularly evident in handling high-resolution images with complex data, achieving errors of fewer than 0.1 mm on tested objects. © 2025, Association for the Advancement of Artificial Intelligence (***). 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

A 3D Implementation of Convolutional Neural Network for Fast Inference

A 3D Implementation of Convolutional Neural Network for Fast...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Narasinga Rao Miniskar Pruek Vanna-Iampikul Aaron Young Sung Kyu Lim Frank Liu Jieun Yoo Corrinne Mills Nhan Tran Farah Fahim Jeffrey S Vetter Computer Science and Mathematic Division Oak Ridge National Laboratory Oak Ridge USA Department of Electrical and Computer Engineering Georgia Institute of Technology Atlanta USA Department of Physics University of Illinois Chicago Chicago USA Scientific Computing Division Fermi National Accelerator Laboratory Chicago USA

Low latency inference has many applications in edge machine learning. In this paper, we present a run-time configurable convolutional neural network (CNN) inference ASIC design for low-latency edge machine learning. By implementing a 5-stage pipelined CNN inference model in a 3D ASIC technology, we demonstrate that the model distributed on two dies utilizing face-to-face (F2F) 3D integration achieves superior performance. Our experimental results show that the design based on 3D integration achieves 43% better energy-delay product when compared to the traditional 2D technology.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Implementation of Preprocessing in Text Summarization Techniques for Indonesian Language Documents Using the Flax T5 Approach

Implementation of Preprocessing in Text Summarization Techni...

引用

International Conference on Cyber and IT Service Management (CITSM)

作者： Arif Ridho Lubis Habibi Ramdani Safitri Irvan Muharman Lubis Al-Khowarizmi Okvi Nugroho Department of Computer Engineering and Informatics Politeknik Negeri Medan Medan Indonesia Department of Mathematic Education Universitas Muhammadiyah Sumatera Utara Medan Indonesia School of Industrial Engineering Telkom University Bandung Indonesia Department of Information Technology Universitas Muhammadiyah Sumatera Utara Medan Indonesia Kulkas IT Medan Indonesia

As information develops, many individuals or communities change their habits in obtaining information, which is usually consumed in large quantities, so that sometimes it is difficult to understand, it becomes easier to get information in summary form, so this research applies the text summarization technique to Indonesian language documents using the Flax T5 model. is the development of transformer-based google. The results of this study are the use of parameters from the Flax T5 model that can affect the performance of text summaries, evidenced by the number of epochs 20, batch size 64 producing ROUGE-1 of 0.24. the ROUGE-2 value is 0.3 and the ROUGE-L value is 0.31, while the epoch value is 60, batch size 128 makes ROUGE-1 0.35, ROUGE-2 value is 0.38 and the ROUGE-L value is 0.36. while the epoch value of 100 and the batch size value of 256 produce a ROUGE-1 value of 0.5, a ROUGE-2 value of 0.54, and a ROUGE-L value of 0.55. However, this model has limitations in text lengths of more than 150 because it will make the training time so long that it is highly recommended not to use text lengths that exceed 150 words.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FNIN: A Fourier Neural Operator-based Numerical Integration Network for Surface-form-gradients

arXiv

引用

arXiv 2025年

Surface-from-gradients (SfG) aims to recover a three-dimensional (3D) surface from its gradients. Traditional methods encounter significant challenges in achieving high accuracy and handling high-resolution inputs, particularly facing the complex nature of discontinuities and the inefficiencies associated with large-scale linear solvers. Although recent advances in deep learning, such as photometric stereo, have enhanced normal estimation accuracy, they do not fully address the intricacies of gradient-based surface reconstruction. To overcome these limitations, we propose a Fourier neural operator-based Numerical Integration Network (FNIN) within a two-stage optimization framework. In the first stage, our approach employs an iterative architecture for numerical integration, harnessing an advanced Fourier neural operator to approximate the solution operator in Fourier space. Additionally, a self-learning attention mechanism is incorporated to effectively detect and handle discontinuities. In the second stage, we refine the surface reconstruction by formulating a weighted least squares problem, addressing the identified discontinuities rationally. Extensive experiments demonstrate that our method achieves significant improvements in both accuracy and efficiency compared to current state-of-the-art solvers. This is particularly evident in handling high-resolution images with complex data, achieving errors of fewer than 0.1 mm on tested objects. Copyright © 2025, The Authors. All rights reserved.

关键词： Surface reconstruction

来源：评论

学校读者我要写书评

暂无评论

Impact of predation in the spread of an infectious disease with time fractional derivative and social behavior

引用

International Journal of Modeling, Simulation, and Scientific Computing 2021年第4期12卷 92-119页

作者： Soufiane Bentout Behzad Ghanbari Salih Djilali Lakshmi Narayan Guin Department of Mathematics and Informatics Belhadj Bouchaib University Center of Ain Temouchent BP 284 RP 46000Algeria Laboratoire d’Analyse Non Linaire et Mathmatiques Appliques Universit de Tlemcen TlemcenAlgeria Department of Engineering Science Kermanshah University of Technology KermanshahIran Department of Biomedical Engineering Faculty of Engineering and Naturel Sciences Bah¸ce¸sehir University 34349 IstanbulTurkey Faculty of Exact and Computer Sciences Mathematic Department Hassiba Benbouali University ChlefAlgeria Department of Mathematics Visva-Bharati Santiniketan-731235 West BengalIndia

The main purpose of this paper is to explore the influence of predation on the spread of a disease developed in the prey population where we assume that the prey has a social *** memory of the prey and the predator measured by the time fractional derivative plays a crucial role in modeling the dynamical response in a predator–prey *** memory can be modeled to articulate the involvement of interacting species by the presence of the time fractional derivative in the considered *** the purpose of studying the complex dynamics generated by the presence of infection and the time-fractional-derivative we split our study into two cases. The first one is devotedto study the effect of a non-selective hunting on the spread of the disease, where the localstability of the equilibria is investigated. Further the backward bifurcation is obtainedconcerning basic reproduction rate of the infection. The second case is for explaining theimpact of selecting the weakest infected prey on the edge of the herd by a predator onthe prevalence of the infection, where the local behavior is scrutinized. Moreover, for thegraphical representation part, a numerical simulation scheme has been achieved usingthe Caputo fractional derivative operator.

关键词： Predator–prey model herd behavior fractional derivative numerical schema backward bifurcation eco-epidemiological model

来源：评论

学校读者我要写书评

暂无评论

Brain Imaging-to-Graph Generation using Adversarial Hierarchical Diffusion Models for MCI Causality Analysis

arXiv

引用

arXiv 2023年

作者： Zuo, Qiankun Tian, Hao Pun, Chi-Man Wang, Hongfei Zhang, Yudong Hong, Jin The School of Information Engineering Hubei University of Economics Hubei Wuhan430205 China The Department of Computer and Information Science University of Macau 519000 China The Department of Orthopaedics and Trauma-tology University of Hong Kong 999077 Hong Kong The School of Computing and Mathematic Sciences University of Leicester LeicesterLE1 7RH United Kingdom The School of Information Engineering Nanchang University Nanchang330031 China

Effective connectivity can describe the causal patterns among brain regions. These patterns have the potential to reveal the pathological mechanism and promote early diagnosis and effective drug development for cognitive disease. However, the current methods utilize software toolkits to extract empirical features from brain imaging to estimate effective connectivity. These methods heavily rely on manual parameter settings and may result in large errors during effective connectivity estimation. In this paper, a novel brain imaging-to-graph generation (BIGG) framework is proposed to map functional magnetic resonance imaging (fMRI) into effective connectivity for mild cognitive impairment (MCI) analysis. To be specific, the proposed BIGG framework is based on the diffusion denoising probabilistic models (DDPM), where each denoising step is modeled as a generative adversarial network (GAN) to progressively translate the noise and conditional fMRI to effective connectivity. The hierarchical transformers in the generator are designed to estimate the noise at multiple scales. Each scale concentrates on both spatial and temporal information between brain regions, enabling good quality in noise removal and better inference of causal relations. Meanwhile, the transformer-based discriminator constrains the generator to further capture global and local patterns for improving high-quality and diversity generation. By introducing the diffusive factor, the denoising inference with a large sampling step size is more efficient and can maintain high-quality results for effective connectivity generation. Evaluations of the ADNI dataset demonstrate the feasibility and efficacy of the proposed model. The proposed model not only achieves superior prediction performance compared with other competing methods but also predicts MCI-related causal connections that are consistent with clinical studies. Copyright © 2023, The Authors. All rights reserved.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Enhanced prediction of defibrillation success in out-of-hospital cardiac arrest using nonlinear ECG features and probabilistic neural network classification

引用

Signal, Image and Video Processing 2025年第8期19卷

作者： Shamsi, Hamed Golkarieh, Alireza Nouri, Houman Dolatabadi, Amirhosein Department of Artificial Intelligence and Data Engineering Atatürk University Yakutiye Erzurum25240 Turkey Department of Computer Science and Engineering Oakland University RochesterMI United States Department of Mathematic University of Sussex Brighton United Kingdom Department of Biomedical Engineering Science and Research Branch Islamic Azad University Tehran Iran

The present study proposes a new approach to predict the outcome of external defibrillation in out-of-hospital cardiac arrest (OHCA) patients using sophisticated nonlinear feature extraction and classification via a probabilistic neural network (PNN) of electrocardiogram (ECG) signals. We analyzed 251 samples of ECG data collected from 195 unsuccessful and 56 successful resuscitations in OHCA patients. Our proposed methodology utilized the following five advanced nonlinear dynamics features: Sample Entropy, Approximate Entropy, Detrended Fluctuation Analysis, Correlation Dimension, and Largest Lyapunov Exponent. These features were selected such that the complex dynamics of ECG signals are not missed by traditional linear analyses and subtle patterns. Then, classification has been performed using the extracted features with a PNN model, which gave great performances with a training accuracy of 97.16% and a testing accuracy of 93.33%. With very good discriminative ability, the model resulted in a high area under the curve (AUC) of 0.98 and 0.96 for both the training and testing sets, respectively. This constitutes a significant advance in the ECG-based prediction of the outcome of defibrillation and could enable better-informed decision-making in emergency cardiac care. Our approach seems clinically applicable given its high accuracy and robustness, though further validation with larger-scale studies would be desirable. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Resuscitation

来源：评论

学校读者我要写书评

暂无评论

The implementation of conductivity data in delineating liquefaction zone in Banda Aceh and Aceh Besar, Indonesia

引用

AIP Conference Proceedings 2024年第1期3082卷

作者： Sabrian Tri Anda Muhammad Syukri Zul Fadhli Mutasar Fiqa Miftahunnisa Physics Department Engineering Faculty Universitas Samudra Langsa Aceh Indonesia Physics Department Faculty of Mathematic and Natural Sciences Univeristas Syiah Kuala Banda Aceh Indonesia Department of Geophysics Engineering Engineering Faculty Univeristas Syiah Kuala Banda Aceh Indonesia Faculty of Computer and Multimedia University of National Islamic Indonesia Bireuen Aceh Indonesia

Earthquake is classified as one of the most disastrous catasthropes regarding casualties and environmental damage. The earthquake itself can act as an agent for another hazard to be occurred such as soil liquefaction phenomenon. The research regarding the delineation of liquefaction zone in Banda Aceh and Aceh Besar has been done by implementing 2-D geoelectric measurement (conductivity data). The study is considered necessary since the tectonic activities in Aceh Province is quite active. The study identified that the two locations of the study area have a high probability of soil liquefaction phenomenon to be occurred indicated by conductivity values of > 0.3 mS/m. The conductivity values of > 0.3 mS/m describe that the layer is in saturated condition due to seawater intrusion. Moreover, the irregular morphology and abrupt change in conductivity values, as well as a saturated layer, lying below a thin layer with low conductivity values, indicate that the area is susceptible to soil liquefaction phenomena.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimization of Time for CAD/CAE Paperless Learning 2

Optimization of Time for CAD/CAE Paperless Learning

引用

2nd Al-Noor International Conference for Science and Technology, NICST 2020

作者： Adhem, Ali Jalghaf, Humam K. Habeeb, Laith J. Deroubi, Nabeh Al Yousif, Ahmed S. Hameed, Firas N. Al-Nahrain Center for Strategic Studies Bagdad Iraq University of Technology Mechanical Engineering Department Bagdad Iraq University of Technology Training and Workshop Center Bagdad Iraq Southeast Community College Computer Aided Design and Drafting United States University of Technology Mechanical Engineering Department Bagdad Iraq University of Anbar College of Education Mathematic Dept. Anbar Iraq

ISBN: (纸本)9781665404327

This study determines an innovative way of teaching, assessing, and examining students devoid of utilizing paper in the lectures, midterms, and last tests of two classes. That is to say, computer-Aided Design and computer-Aided engineering, for third and fourth-year students, respectively. Paperless learning is the name for this way, in which computers are utilized interactively to teach and assess students and take tests. The students are involved interactively in studying scientific lessons;computer software is used to capitalize on their potential to develop learning and help students do reproductions to resolve different engineering exercises. The outcomes exhibited that the students deliver a significant consideration to the lessons utilized in the two classes. Furthermore, the rating procedure is more proficient than the usual methods. The review showed that the students tend to be positive in evaluating the lessons and professors. Additionally, the significance of the other profits of the paperless method is offered. © 2020 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

A normalized cross-correlation convolutional neural network (CNN-NCC) for exemplar-based object detection

引用

AIP Conference Proceedings 2023年第1期2734卷

作者： Yaya Wihardi Winda M. Kristy Erlangga Arjon Turnip Intan N. Yulita Endroyono Computer Science Education Department Faculty of Mathematic and Natural Science Education Universitas Pendidikan Indonesia Bandung Indonesia Artificial Intelligence and Big Data Research Center Padjadjaran University Sumedang Indonesia Department of Electrical Engineering Sepuluh Nopember Institute of Technology Surabaya Indonesia

Object detection in dynamic environments, such as CCTV surveillance systems and robotics, requires a special approach. New objects that have not been previously trained to the model are often captured by sensors, and it is very important to recognize them in the future. One approach that can be used to solve this problem is an exemplar-based approach. On this study, we try to build an exemplar-based object detector based on Convolutional Neural Network with Normalized Cross Correlation matching. Our proposed method did not need to re-train the model to identify unknown objects that had not been acknowledged by the model during the training stage. The results show that the model achieves 0.449 mAP on the COCO dataset.

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：