检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Kim, Junho Choi, Changwoon Jang, Hojun Kim, Young Min Dept. of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence and INMC Seoul National University Korea Republic of

We introduce LDL, a fast and robust algorithm that localizes a panorama to a 3D map using line segments. LDL focuses on the sparse structural information of lines in the scene, which is robust to illumination changes and can potentially enable efficient computation. While previous line-based localization approaches tend to sacrifice accuracy or computation time, our method effectively observes the holistic distribution of lines within panoramic images and 3D maps. Specifically, LDL matches the distribution of lines with 2D and 3D line distance functions, which are further decomposed along principal directions of lines to increase the expressiveness. The distance functions provide coarse pose estimates by comparing the distributional information, where the poses are further optimized using conventional local feature matching. As our pipeline solely leverages line geometry and local features, it does not require costly additional training of line-specific features or correspondence matching. Nevertheless, our method demonstrates robust performance on challenging scenarios including object layout changes, illumination shifts, and large-scale scenes, while exhibiting fast pose search terminating within a matter of milliseconds. We thus expect our method to serve as a practical solution for line-based localization, and complement the well-established point-based paradigm. The code for LDL is available through the following link: https://***/82magnolia/ panoramic-localization. © 2023, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Stretchable Textile Bands for Ambulatory Electrocardiogram and Oximetry

IEEE Journal on Flexible Electronics

引用

IEEE Journal on Flexible Electronics 2022年第4期1卷 214-222页

作者： Liu, Jiaxi Le, Thy Sadatian, Ellie Yao, Lulu Garudadri, Harinath Dunphy, Stephen A. McClung, Christian Ng, Tse Nga University of California at San Diego Department of Electrical and Computer Engineering La Jolla CA92093 United States University of California at San Diego Materials Science and Engineering Program La Jolla CA92093 United States Qualcomm Institute University of California at San Diego La Jolla CA92093 United States Cb Innovations Llc Rancho Santa FeCA92067 United States

This work presents a vital sign monitoring interface combining electrocardiogram (ECG) and reflective photoplethysmography (PPG) acquisition on stretchable kinesiology tapes (KTs). The integrated textile bands are less prone to electrode misplacement and deliver high-quality diagnostic ECG signals with dry electrodes. The reflective PPG measured over the subclavian artery is used to track blood oxygen saturation, and in combination with ECG, enables the determination of pulse transit time and inference of blood pressure. This multimodal interface improves clinical workflows because it is easy to don, reduces motion artifacts, and facilitates more efficient and accurate diagnosis in emergency settings. © 2022 IEEE.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Bayesian Parameter Inference for Partially Observed Stochastic Volterra Equations

arXiv

引用

arXiv 2023年

作者： Jasra, Ajay Ruzayqat, Hamza Wu, Amin Applied Mathematics & Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955 Saudi Arabia

In this article we consider Bayesian parameter inference for a type of partially observed stochastic Volterra equation (SVE). SVEs are found in many areas such as physics and mathematical finance. In the latter field they can be used to represent long memory in unobserved volatility processes. In many cases of practical interest, SVEs must be time-discretized and then parameter inference is based upon the posterior associated to this time-discretized process. Based upon recent studies on time-discretization of SVEs (e.g. [24]) we use Euler-Maruyama methods for the afore-mentioned discretization. We then show how multilevel Markov chain Monte Carlo (MCMC) methods [19] can be applied in this context. In the examples we study, we give a proof that shows that the cost to achieve a 4 mean square error (MSE) of O(ϵ2), ϵ > 0, is O(ϵ− 4/2H+1 ), where H is the Hurst parameter. If one uses a single level MCMC method then the cost is O(ϵ−2(2H+3)/ 2H+1 ) to achieve the same MSE. We illustrate these results in the context of state-space and stochastic volatility models, with the latter applied to real *** Codes 62F15, 62M20, 60J10, 60J22, 65C40 © 2023, CC BY.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

External Lightning Protection System modeling on photovoltaic using ATP Draw

External Lightning Protection System modeling on photovoltai...

引用

International Conference on electrical engineering, computer Science and Informatics (EECSI)

作者： Rohana Rohana Surya Hardi Nasaruddin Nasaruddin Yuwaldi Away Doctoral Program. School of Engineering Universitas Syiah Kuala Banda Aceh Indonesia and Universitas Muhammadiyah Sumatera Utara Indonesia Dept. of Electrical Engineering Universitas Sumatera Utara Medan Indonesia Dept. of Electrical and Computer Engineering Universitas Syiah Kuala Banda Aceh Indonesia Dept. of Electrical and Computer Universitas Syiah Kuala Banda Aceh Indonesia

Solar energy is environmentally friendly energy with a very large energy source. Indonesia is also exploring and implementing renewable energy options including solar panels as a viable and long-term alternative to green technology. On the other hand, to get sufficient solar radiation, solar panel systems are generally located in open areas such as hillsides, deserts, and water surfaces, making them more vulnerable to lightning strikes. Lightning is one of the main threats to PLTS, a lot of damage to solar panels is caused by lightning. Direct and indirect lightning strikes have great potential to affect the entire solar panel farm system. The main objective of this study is to evaluate Lightning Protection System (LPS) modeling for network-connected solar panel (PV) farm systems using the ATP-EMTP software. Field observations and simulation tests are used to determine the position of the LPS and its installation structure. The rolling ball method is used to see the installed LPS protection radius. The results show that it is necessary to add LPS to the PV field. It was found that there was one additional LPS needed on the north side of the PV field, because that position was not protected by the radius of the lightning rod. From the simulation results, it was found that the LPS installation structure was good, because the results from the modeling of the isolated LPS lightning protection system were better to use than without installing LPS.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AV Navigation in 3D Urban Environments with Curriculum-based Deep Reinforcement Learning

AV Navigation in 3D Urban Environments with Curriculum-based...

引用

International Conference on Unmanned Aircraft Systems (ICUAS)

作者： Kevin Braathen de Carvalho Iure Rosa L. de Oliveira Alexandre S. Brandão Department of Electrical Engineering Graduate Program in Computer Science Nucleus of Specialization in Robotics Universidade Federal de Viçosa Viçosa MG Brazil Department of Electrical Engineering Federal University of Viçosa Viçosa MG Brazil

Unmanned Aerial Vehicles (UAVs) are widely used in various applications, from inspection and surveillance to transportation and delivery. Navigating UAVs in complex 3D environments is a challenging task that requires robust and efficient decision-making algorithms. This paper presents a novel approach to UAV navigation in 3D environments using a Curriculum-based Deep Reinforcement Learning (DRL) approach. The proposed method utilizes a deep neural network to model the UAV’s decision-making process and to learn a mapping from the state space to the action space. The learning process is guided by a reinforcement signal that reflects the performance of the UAV in terms of reaching its target while avoiding obstacles and with energy efficiency. Simulation results show that the proposed method has a positive trade off when compared to the baseline algorithm. The proposed method was able to perform well in environments with a state space size of 22 millions, allowing the usage in big environments or in maps with high resolution. The results demonstrate the potential of DRL for enabling UAVs to operate effectively in complex environments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DAFA: DISTANCE-AWARE FAIR ADVERSARIAL TRAINING

arXiv

引用

arXiv 2024年

作者： Lee, Hyungyu Lee, Saehyung Jang, Hyemi Park, Junsung Bae, Ho Yoon, Sungroh Electrical and Computer Engineering Seoul National University Korea Republic of Department of Cyber Security Ewha Womans University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

The disparity in accuracy between classes in standard training is amplified during adversarial training, a phenomenon termed the robust fairness problem. Existing methodologies aimed to enhance robust fairness by sacrificing the model's performance on easier classes in order to improve its performance on harder ones. However, we observe that under adversarial attacks, the majority of the model's predictions for samples from the worst class are biased towards classes similar to the worst class, rather than towards the easy classes. Through theoretical and empirical analysis, we demonstrate that robust fairness deteriorates as the distance between classes decreases. Motivated by these insights, we introduce the Distance-Aware Fair Adversarial training (DAFA) methodology, which addresses robust fairness by taking into account the similarities between classes. Specifically, our method assigns distinct loss weights and adversarial margins to each class and adjusts them to encourage a trade-off in robustness among similar classes. Experimental results across various datasets demonstrate that our method not only maintains average robust accuracy but also significantly improves the worst robust accuracy, indicating a marked improvement in robust fairness compared to existing methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Gradient Alignment with Prototype Feature for Fully Test-time Adaptation

arXiv

引用

arXiv 2024年

作者： Shin, Juhyeon Lee, Jonghyun Lee, Saehyung Park, Minjun Lee, Dongjun Hwang, Uiwon Yoon, Sungroh Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Department of Electrical and Computer Engineering Seoul National University Korea Republic of Division of Digital Healthcare Yonsei University Korea Republic of

In context of Test-time Adaptation(TTA), we propose a regularizer, dubbed Gradient Alignment with Prototype feature (GAP), which alleviates the inappropriate guidance from entropy minimization loss from misclassified pseudo label. We developed a gradient alignment loss to precisely manage the adaptation process, ensuring that changes made for some data don’t negatively impact the model’s performance on other data. We introduce a prototype feature of a class as a proxy measure of the negative impact. To make GAP regularizer feasible under the TTA constraints, where model can only access test data without labels, we tailored its formula in two ways: approximating prototype features with weight vectors of the classifier, calculating gradient without back-propagation. We demonstrate GAP significantly improves TTA methods across various datasets, which proves its versatility and effectiveness. Copyright © 2024, The Authors. All rights reserved.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Joint Source Decomposition of Concurrent EEG-FMRI Data in Epilepsy and Control Groups 22

Joint Source Decomposition of Concurrent EEG-FMRI Data in Ep...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Harding, J. Mason Pourmotabbed, Haatef Li, Yamin Rogge-Obando, Kimberly Wang, Kate Goodale, Sarah E. Wang, Shiyu Bibro, Camden Allee, Bergen Martin, Caroline Morgan, Victoria L. Englot, Dario J. Chang, Catie Departments of Electrical and Computer Engineering Vanderbilt University United States Departments of Biomedical Engineering Vanderbilt University United States Departments of Computer Science Vanderbilt University United States Neuroscience Graduate Program Vanderbilt University United States Departments of Radiology Vanderbilt University Medical Center United States Departments of Neurological Surgery Vanderbilt University Medical Center United States Vanderbilt Memory and Alzheimer's Center Vanderbilt University Medical Center United States

ISBN: (纸本)9798331520526

EEG and fMRI are complementary, noninvasive technologies for investigating human brain function. These modalities have been used to uncover large-scale functional networks and their disruptions in clinical populations. Given the high temporal resolution of EEG and high spatial resolution of fMRI, integrating these modalities can provide a more holistic understanding of brain activity. This work explores a multimodal source decomposition technique for extracting shared modes of temporal variation between fMRI BOLD signals and EEG spectral power fluctuations in the resting state. The resulting components are then compared between patients with focal epilepsy and controls, revealing multimodal network differences between groups. © 2025 IEEE.

关键词： Brain mapping

来源：评论

学校读者我要写书评

暂无评论

Global distribution of forest classes and leaf biomass for use as alternative foods to minimize malnutrition

World Food Policy

引用

World Food Policy 2021年第2期7卷 128-146页

作者： Fist, Tim Adesanya, Adewale A. Denkenberger, David Pearce, Joshua M. Alliance to Feed the Earth in Disasters (ALLFED) Environmental and Energy Program Social Science Department Michigan Technological University Department of Mechanical Engineering University of Alaska Fairbanks Department of Electrical and Computer Engineering Western University

Due to the ready availability of tree leaves in many geographies, the alternative food of leaf concentrate currently has the potential to alleviate hunger in over 800 million people. It is therefore potentially highly impactful to determine the edibility of leaf concentrates, which are in the same regions as the world's most undernourished populations. Unfortunately, the toxicity of leaf concentrate for most common tree leaf types has not been screened and the cost of doing so demands a prioritization. This preliminary study explores this potential solution to world hunger by finding the forest classes most likely to offer proximate access to the world's hungry, thus providing the basis for a prioritized list of leaf types to screen for toxicity. Specifically, this study describes a novel methodology for mapping available green leaf biomass and corresponding forest classes (e.g., tropical moist deciduous forest), and their spatial relationship to the global distribution of people who are underweight. These results will be useful for developing a targeted list of tree species to conduct leaf toxicity analysis on, in the interest of developing leaves as an alternative food source for both current malnutrition problems and global catastrophic scenarios. © 2021 Policy Studies Organization.

关键词： agriculture edible leaves forests GIS global catastrophic risk hunger

来源：评论

学校读者我要写书评

暂无评论

More Efficient Training Strategy to Leverage Neurons in Neural Network

More Efficient Training Strategy to Leverage Neurons in Neur...

引用

IEEE International Symposium on Industrial Electronics (ISIE)

作者： Cheng-Fu Liou Yi-Cheng Yu Research Labs (MZAIR Labs) Meta Zen Artificial Intelligence Tainan Taiwan Graduate Degree Program of College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu Taiwan

ISBN: (数字)9798350394085

ISBN: (纸本)9798350394092

Comparing the structures of current neural networks and the biological brain, it can be observed that many mechanisms correspond to each other according to their functions. From the perspective of the biological brain, different components have different long-term or short-term memory effects, such as the hippopotamus. However, current neural networks do not pay much attention to this aspect. In this work, we incorporated many hypotheses or theories inspired by the territory of neurobiology to retain the learned knowledge. To start with, we follow the inspiration of the synaptic homeostasis hypothesis (SHY) [1] and add an additional training stage to the training process of the model, which can ensure that only the most important information remains intact and the insignificant synapse can be pruned. In other words, we divide the overall training process into two learning stages: synaptogenesis and synaptic sparsifying. We experimentally demonstrate that our novel learning strategy significantly outperforms the traditional solution in training a sequence of tasks at different times on three public datasets, which supports that the proposed method is more efficient for resource-limited edge devices.

关键词： Training Industrial electronics Neurons Homeostasis Task analysis Biological neural networks Synapses

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：