检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Bidollahkhani, Michael Atasoy, Ferhat Abdellatef, Hamdan Department of Computer Engineering Institute of Graduate Studies Karabuk University Karabuk Turkey Gesellschaft für Wissenschaftliche DatenverarbeitungmbH Göttingen Germany Department of Electrical and computer engineering Lebanese American University Byblos Lebanon

We present LTC-SE, an improved version of the Liquid Time-Constant (LTC) neural network algorithm originally proposed by Hasani et al. in 2021. This algorithm unifies the Leaky-Integrate-and-Fire (LIF) spiking neural network model with Continuous-Time Recurrent Neural Networks (CTRNNs), Neural Ordinary Differential Equations (NODEs), and bespoke Gated Recurrent Units (GRUs). The enhancements in LTC-SE focus on augmenting flexibility, compatibility, and code organization, targeting the unique constraints of embedded systems with limited computational resources and strict performance requirements. The updated code serves as a consolidated class library compatible with TensorFlow 2.x, offering comprehensive configuration options for LTCCell, CTRNN, NODE, and CTGRU classes. We evaluate LTC-SE against its predecessors, showcasing the advantages of our optimizations in user experience, Keras function compatibility, and code clarity. These refinements expand the applicability of liquid neural networks in diverse machine learning tasks, such as robotics, causality analysis, and time-series prediction, and build on the foundational work of Hasani et al. © 2023, CC BY-NC-ND.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

Harnessing Dataset Cartography for Improved Compositional Generalization in Transformers

arXiv

引用

arXiv 2023年

作者： Ince, Osman Batur Zeraati, Tanin Yagcioglu, Semih Yaghoobzadeh, Yadollah Erdem, Erkut Erdem, Aykut Koç University KUIS AI Center Turkey Koç University Computer Engineering Department Turkey University of Tehran Iran Hacettepe University Computer Engineering Department Turkey Tehran Institute for Advanced Studies Khatam University Iran

Neural networks have revolutionized language modeling and excelled in various downstream tasks. However, the extent to which these models achieve compositional generalization comparable to human cognitive abilities remains a topic of debate. While existing approaches in the field have mainly focused on novel architectures and alternative learning paradigms, we introduce a pioneering method harnessing the power of dataset cartography (Swayamdipta et al., 2020). By strategically identifying a subset of compositional generalization data using this approach, we achieve a remarkable improvement in model accuracy, yielding enhancements of up to 10% on CFQ and COGS datasets. Notably, our technique incorporates dataset cartography as a curriculum learning criterion, eliminating the need for hyperparameter tuning while consistently achieving superior performance. Our findings highlight the untapped potential of dataset cartography in unleashing the full capabilities of compositional generalization within Transformer models. Our code is available at https://***/cyberiada/ cartography-for-compositionality. Copyright © 2023, The Authors. All rights reserved.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Systematic Stress Detection in CNN Application 10

Systematic Stress Detection in CNN Application

引用

10th International Conference on Reliability, Infocom Technologies and Optimization ,Trends and Future Directions, ICRITO 2022

作者： Mohapatra, Srikanta Kumar Kishore Kanna, R. Arora, Ginni Sarangi, Prakash Kumar Mohanty, Jayashree Sahu, Premananda Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Vels Institute of Science Technology & Advanced Studies Department of Biomedical Engineering Amity University Department of Computer Science and Engineering New Delhi Noida India Hyderabad India Odisha Bhubaneswar India Srm Institute of Science and Technology Srmist UP Ghaziabad India

ISBN: (数字)9781665474337

ISBN: (纸本)9781665474337

Emotional strain is referred to as stress. Both our mental health and the mental health of others around us may be affected. While anxiety is a normal response to stress that may be frightening, it can also cause panic attacks. Everyone needs to address these mental health problems. The method we use to identify stress and anxiety in a person using vocal/audio information is described in this study. We have created a deep neural network model for stress and anxiety detection. Here, Kaggle audio datasets with 7 different emotions―joy, fear, disgust, neutral, sorrow, surprise, and anger―are taken into consideration. To train and evaluate classification algorithms like CNN, these audio datasets are employed. The audio is then pre-processed using acoustic feature extraction, and CNN is used to classify it, providing accuracy based on those seven emotions. This allows us to foretell if the individual is stressed out or anxious. © 2022 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Navigating the Edge: UAS Boundary Tracing for Efficient Volcanic Plume Monitoring

Navigating the Edge: UAS Boundary Tracing for Efficient Volc...

引用

IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)

作者： John Ericksen Abir Islam Carter Frost Kevin Aubert G. Matthew Fricke Varsha Dani Rafael Fierro Tobias Fischer Scott Nowicki Jared Saia Melanie Moses Department of Computer Science University of New Mexico Department of Electrical Engineering University of New Mexico Center for Advanced Research Computing University of New Mexico Computer Science Department Rochester Institute of Technology Earth and Planetary Sciences University of New Mexico Santa Fe Institute Santa Fe

ISBN: (数字)9798331510954

ISBN: (纸本)9798331510961

We present the implementation and validation of Sketch, an algorithm that uses two Unpiloted Aerial Systems (UASs) to trace the boundary of volcanic plumes. Sketchguarantees asymptotically optimal flight distance and turning by maintaining a sandwich invariant where one UAS stays inside the plume boundary (defined by a $\textbf{CO}_{2}$ concentration threshold) and the other UAS stays outside. The UASs adjust their flight paths based on real-time CO 2 measurements to maintain this invariant. This paper details the implementation of Sketchon a real-world UAS platform, the Dragonfly drone. We evaluate the efficacy of Sketchthrough extensive testing in physics-based simulations and real-world outdoor environments using virtual plumes. The algorithm is compared to a single-UAS baseline algorithm called Zigzag. Results show that Sketchmeets the expectations set by theory, and it is more efficient than Zigzag, achieving shorter flight paths, less turning, and faster mapping times. While Zigzagexhibits slightly higher accuracy in estimating plume area and boundary, Sketchoffers a more efficient real-time volcanic plume monitoring approach, especially in time-sensitive situations. These results demonstrate the feasibility and efficacy of Sketchfor real-time volcanic plume monitoring, paving the way for accurate $\textbf{CO}_2$ emission estimation in hazardous and challenging environments.

关键词： Accuracy Navigation Volcanic activity Estimation Turning Real-time systems Safety Security Monitoring Testing

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning Approach on Bacteria Classification from Microscopic Images 5

Transfer Learning Approach on Bacteria Classification from M...

引用

5th International Conference on Contemporary Computing and Informatics, IC3I 2022

作者： Singh, Anupam Kumar, Abhishek Salman, Hayder M. Rawat, Navneet Jain, Sanjiv Kumar Rao, Annam Takshitha Shobhit University UP Meerut India Al-Turath Universiy College Department of Computer Science Baghdad10013 Iraq Graphic Era Deemed to Be University Department of Management Studies Uttarakhand Dehradun India Medi-Caps University Electrical Engineering Department Indore India Symbiosis Institute of Technology SIT Department of Computer Science and Engineering Maharashtra Pune India

ISBN: (纸本)9798350398267

The ability to identify and categorize bacteria is crucial in modern medicine for disease diagnosis, infection treatment, and epidemic investigation. However, manually identification and categorization of bacteria requires a lot of time and effort from humans. As technology has progressed, computer systems-based techniques are now doing the duty of identifying images captured by digital electron microscopes. On top of that, modern Deep Learning (DL) methods have shown remarkable improvement in the area of image classification. In this research, we explore a method for using a DL model to automate the identification and categorization of bacteria. To develop the DL model, we used a dataset consisting of more than 600 images of 33 distinct bacteria taken with a microscope and the 'transfer learning' technique. GoogLeNet and AlexNet are two examples of transfer learning models used in this research. The DL classification accuracy was evaluated using 20% randomly selected and isolated images from the dataset. Experimental findings of prediction obtained an accuracy of roughly 98.67% by GoogLeNet, and both transfer learning models recognized and classified all 33 bacterial species with better success rates. © 2022 IEEE.

关键词： Bacteria

来源：评论

学校读者我要写书评

暂无评论

Experimental demonstration of a flexible-grid 1×(2×3)mode-and wavelength-selective switch using silicon microring resonators and counter-tapered couplers

引用

Chinese Optics Letters 2024年第1期22卷 52-58页

作者：孔德军鲁昊汪鹏君符强戴世勋陈伟伟王栎沣张波豪马凌霄李军戴庭舸杨建义 Faculty of Electrical Engineering and Computer Science Ningbo UniversityNingbo 315211China College of Electrical and Electronic Engineering Wenzhou UniversityWenzhou 325035China College of Science and Technology Ningbo UniversityNingbo 315300China Laboratory of Infrared Materials and Devices The Research Institute of Advanced TechnologiesNingbo UniversityNingbo 315211China Department of Information Science and Electronics Engineering and Cyrus Tang Center for Sensor Materials and Applications Zhejiang UniversityHangzhou 310027China

A flexible-grid 1×(2×3)mode-and wavelength-selective switch which comprises counter-tapered couplers and silicon microring resonators has been proposed,optimized,and demonstrated experimentally in this *** carefully thermally tuning phase shifters and silicon microring resonators,mode and wavelength signals can be independently and flexibly conveyed to any one of the output ports,and different bandwidths can be generated as *** particle swarm optimization algorithm and finite difference time-domain method are employed to optimize structural parameters of the twomode(de)multiplexer and crossing *** bandwidth-tunable wavelength-selective optical router composed of12 microring resonators is studied by taking advantage of the transfer matrix *** results show that,for the fabricated module,cross talk less than-10.18 dB,an extinction ratio larger than 17.41 d B,an in-band ripple lower than0.79 dB,and a 3-dB bandwidth changing from 0.38 to 1.05 nm are obtained,as the wavelength-channel spacing is 0.40 *** corresponding response time is measured to be 13.64μs.

关键词： integrated optics optical waveguide mode-and wavelength-selective switch

来源：评论

学校读者我要写书评

暂无评论

A Novel Deep Learning Approach for Capturing Time Series Dependencies and Improving Short-term Weather Forecasting 7

A Novel Deep Learning Approach for Capturing Time Series Dep...

引用

7th International Conference on Image Information Processing, ICIIP 2023

作者： Johri, Shiva Divyajyothi, M.G. Anitha, S. Rani, M. Sandya Murari, Thejovathi Shirisha, N. Oriental College of Management Oriental Group of Institutes Department of Management Madhya Pradesh Bhopal462022 India University of Technology And Applied Sciences - Al Mussanah Department of Information Technology Al Muladha Muscat Oman Vels Institute of Science Technology And Advanced Studies Department of Computer Science Engineering Pallavaram Krishnapuram Tamil Nadu Chennai India St. Martin's Engineering College Secunderabad Telangana Dhulapally500100 India Acharya Nagarjuna University Department of Computer Science And Engineering Andhra Pradesh Guntur India Mlr Institute of Technology Department of Cse Hyderabad Dundigal India

ISBN: (纸本)9798350371406

Weather forecasting is a critical and challenging task that requires accurate predictions based on historical data and intricate dependencies between time series. Traditional neural networks, such as Back Propagation through Time (BPTT) trained RNNs, struggle to effectively capture these dependencies, leading to suboptimal performance in weather forecasting. In this study, we suggest a novel Integrated-Stack based Bi-LSTM Model, which associations the strengths of LSTM and Bi-LSTM networks, to address these limitations and enhance the accuracy of short-term weather predictions. The primary objective of this research is to develop a versatile deep learning model capable of maintaining long-term dependances and reading any length of sequence, while simultaneously providing real-time short-term weather forecasts. To achieve this, the Integrated-Stack based Bi-LSTM Model is designed to exploit the versatility and non-linear adaptive processing ability of neural networks. The proposed model's architecture involves the integration of LSTM and Bi-LSTM layers. By incorporating both forward and reverse direction processing, the model gains a comprehensive understanding of complex patterns present in time-series data. This bidirectional approach enables the model to capture dependencies effectively, which is essential for accurate weather predictions. To estimate the presentation of the Integrated-Stack based Bi-LSTM Model, we conducted experiments using real-world weather data. We compared the model's predictive capabilities against traditional LSTM models and other state-of-the-art weather forecasting methods. The assessment metrics included accuracy, root mean square error (RMSE), and mean absolute error (MAE). © 2023 IEEE.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Early Retinal Degeneration Detection with Deep Faster CNNs

Early Retinal Degeneration Detection with Deep Faster CNNs

引用

International Conference on Advances in Information Technology (ICAIT)

作者： K. Priyadharshini Sindhusaranya Balraj JV Rama Kumar Deepak A. Vidhate Thilakavathy. P P. Vanathi Department of Computing Technologies School of Computing SRM Institute of Science and Technology Chennai India Department of Computer Science and Engineering Sona College of Technonolgy Salem India Computer Science and Engineering SRKR Engineering College Bhimavaram Andhra Pradesh Department of Information Technology Dr. Vithalrao Vikhe Patil College of Engineering Ahmednagar Maharashtra Computer science and Engineering Vels Institute of Science Technology and Advanced Studies Chennai Department of Information Technology Adhiyamaan College of Engineering Hosur

ISBN: (数字)9798350383867

ISBN: (纸本)9798350383874

The issue of retinogenesis probably constitutes a major public health threat which gives the reason to carry out its early detection for intervention purposes to be efficient. Throughout this investigation, we suggest a new technique which uses the faster CNNs for early detection of retina degradation. We set up a diverse data collection where OCT(optical coherence tomography), fundus photography and angiography modality were included. This collection covered the various stages of degeneration. The dataset was prepared for instance with resizing, normalizing, and data augmentation were used to improve share quality and diversity. Deeper and faster networks that were developed and trained using the backpropagation technique performed well and further optimized using Adam, with high accuracy of 94 in the test dataset. 7%. The strength of our results can be seen through evaluation metrics in detail like confusion matrices and classification reports. The model can distinguish various cases of retinal degeneration accurately which proves to have high precision and recall for mild and normal degeneration. Interpolation analysis of the model using Grad-CAM and occlusion sensitivity, giving the clue about the model's decision making process, setting a precedent for clinical interpretation. The enhanced cross-validated model on an independent dataset also arrays the model in a representational where it will be able to generalize in a diverse population and in different imaging conditions. The systematic approach which proved to be more accurate and resilient than the other techniques was the one which was compared to the mentioned ones. In conclusion, our study shows the possibility for application of the Deep Faster CNNs in identifying the early signs of retinal degeneration; the tool therefore becomes an important tool for enhancing patient outcomes and prevention of vision loss in the world at large.

关键词： Training Analytical models Technological innovation Accuracy Systematics Coherence Tomography Retina Data models Diseases

来源：评论

学校读者我要写书评

暂无评论

Modified Kelvin-Voigt fractional derivative model for viscoelasticity measurement in optical coherence elastography

引用

Chinese Optics Letters 2025年第1期23卷 93-101页

作者： Chenming Yang Zhongliang Li Nan Nan Teng Liu Yaoli Luo Xiangzhao Wang Advanced Light Source and System R&D Center Department of Advanced Optical and Microelectronic EquipmentShanghai Institute of Optics and Fine MechanicsChinese Academy of SciencesShanghai 201800China Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of SciencesBeijing 100049China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China State Key Laboratory of Extreme Photonics and Instrumentation College of Optical Science and EngineeringZhejiang UniversityHangzhou 310027China

Optical coherence elastography(OCE)can quantitatively obtain the viscoelasticity of tissues using rheological models and is widely applied to the clinical diagnosis of ***,commonly used rheological models in OCE do not account for the distinctive dependence of high-frequency storage and loss moduli on frequency in tissues,which results in the rheological models failing to accurately measure the viscoelastic properties of *** this paper,a modified Kelvin-Voigt fractional derivative model is presented based on the power-law behavior of soft tissues and the dependence of high-frequency complex shear modulus on frequency in living *** the rheometer and OCE tests,the modified model can provide the prediction of the power-law relationship between the low-frequency shear viscosity and frequency;compared with the Kelvin-Voigt and Kelvin-Voigt fractional derivative models,the modified model has a higher goodness-of-fit(accuracy>96%)for the high-frequency storage moduli of gelatin ***,the proposed model can reduce the root mean square error of fit by approximately 83%for the high-frequency(1–128 kHz)storage modulus of the polydimethylsiloxane phantoms obtained from publicly available ***,the modified model accurately predicts the mechanical properties of biomimetic materials over a wide frequency range,with the potential to more accurately reflect pathological changes in tissues.

关键词： optical coherence tomography optical coherence elastography viscoelasticity rheological model

来源：评论

学校读者我要写书评

暂无评论

Terahertz plasmonic resonances in coplanar graphene nanoribbon structures

arXiv

引用

arXiv 2024年

作者： Ryzhii, V. Tang, C. Otsuji, T. Ryzhii, M. Shur, M.S. Research Institute of Electrical Communication Tohoku University Sendai980-8577 Japan Frontier Research Institute for Interdisciplinary Studies Tohoku University Sendai980-8577 Japan Department of Computer Science and Engineering University of Aizu Aizu-Wakamatsu965-8580 Japan Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute TroyNY12180 United States

We analyze plasmonic oscillations in the coplanar graphene nanoribbon (GNR) structures induced by the applied terahertz (THz) signals and calculate the GNR impedance. The plasmonic oscillations in the GNR structures are associated with the electron and hole inductances and the lateral inter-GNR capacitance. A relatively low inter-GNR capacitance enables the resonant excitation of the THz plasmonic oscillations in the GNR structures with long GNRs. The GNR structures under consideration can be used in different THz devices as the resonant structures incorporated in THz detectors, THz sources using resonant-tunneling diodes, photomixers, and surface acoustic wave sensors. © 2024, CC BY.

关键词： Graphene

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：