检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Murali, Nihal Puli, Aahlad Yu, Ke Ranganath, Rajesh Batmanghelich, Kayhan Intelligent Systems Program University of Pittsburgh United States Department of Computer Science New York University United States Department of Electrical and Computer Engineering Boston University United States

Deep Neural Networks (DNNs) are prone to learning spurious features that correlate with the label during training but are irrelevant to the learning problem. This hurts model generalization and poses problems when deploying them in safety-critical applications. This paper aims to better understand the effects of spurious features through the lens of the learning dynamics of the internal neurons during the training process. We make the following observations: (1) While previous works highlight the harmful effects of spurious features on the generalization ability of DNNs, we emphasize that not all spurious features are harmful. Spurious features can be "benign" or "harmful" depending on whether they are "harder" or "easier" to learn than the core features for a given model. This definition is model and dataset dependent. (2) We build upon this premise and use instance difficulty methods (like Prediction Depth (Baldock et al., 2021)) to quantify "easiness" for a given model and to identify this behavior during the training phase. (3) We empirically show that the harmful spurious features can be detected by observing the learning dynamics of the DNN’s early layers. In other words, easy features learned by the initial layers of a DNN early during the training can (potentially) hurt model generalization. We verify our claims on medical and vision datasets, both simulated and real, and justify the empirical success of our hypothesis by showing the theoretical connections between Prediction Depth and information-theoretic concepts like V-usable information (Ethayarajh et al., 2021). Lastly, our experiments show that monitoring only accuracy during training (as is common in machine learning pipelines) is insufficient to detect spurious features. We, therefore, highlight the need for monitoring early training dynamics using suitable instance difficulty metrics. © 2023, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Passive Cancellation Method for Common-Mode Noise Based on a Nonlinear Model

Passive Cancellation Method for Common-Mode Noise Based on a...

引用

Asia-Pacific Symposium on Electromagnetic Compatibility (APEMC)

作者： Yeu-Torng Yau Thanh-Phu Luu Department of Ph.D. Program Prospective Technology of Electrical Engineering and Computer Science National Chin-Yi University of Technology Taichung Taiwan Department of Electrical Engineering National Chin-Yi University of Technology Taichung Taiwan

ISBN: (数字)9784885523472

ISBN: (纸本)9798350349498

AC-DC converters almost always require common-mode (CM) filters to restrict conducting their electromagnetic interference (EMI) emissions to the power grid and ensure compliance with EMI certification requirements. These CM filters, which are based on the well-known LC lowpass filter, are typically bulky and expensive. Hence, a method for passive $\mathbf{C M}$ current cancellation using a nonlinear model is proposed herein. This method is entirely composed of passive components and offers advantages in terms of cost and system reliability. Compared to conventional passive LC filters, the number of magnetic components remains unchanged. This study proposes the derivation of a nonlinear model to estimate the filter performance using measured data from an actual CM choke; however, the proposed method does not require the mathematical model of a CM choke.

关键词： Magnetic separation Electromagnetic interference Low-pass filters Electromagnetic compatibility Noise cancellation Mathematical models Power grids

来源：评论

学校读者我要写书评

暂无评论

Fully Geometric Panoramic Localization

arXiv

引用

arXiv 2024年

作者： Kim, Junho Jeong, Jiwon 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 Dept. of Electrical Engineering Stanford University United States

We introduce a lightweight and accurate localization method that only utilizes the geometry of 2D-3D lines. Given a pre-captured 3D map, our approach localizes a panorama image, taking advantage of the holistic 360◦ view. The system mitigates potential privacy breaches or domain discrepancies by avoiding trained or hand-crafted visual descriptors. However, as lines alone can be ambiguous, we express distinctive yet compact spatial contexts from relationships between lines, namely the dominant directions of parallel lines and the intersection between nonparallel lines. The resulting representations are efficient in processing time and memory compared to conventional visual descriptor-based methods. Given the groups of dominant line directions and their intersections, we accelerate the search process to test thousands of pose candidates in less than a millisecond without sacrificing accuracy. We empirically show that the proposed 2D-3D matching can localize panoramas for challenging scenes with similar structures, dramatic domain shifts or illumination changes. Our fully geometric approach does not involve extensive parameter tuning or neural network training, making it a practical algorithm that can be readily deployed in the real world. Project page including the code is available through this link: https://***/fgpl/. © 2024, CC BY.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Real-Time Computation of E-Field in Transcranial Magnetic Stimulation for Neuronavigation and Optimization

Real-Time Computation of E-Field in Transcranial Magnetic St...

引用

US National Committee of URSI National Radio Science Meeting (USNC-URSI NRSM)

作者： Nahian I. Hasan Moritz Dannhauer Dezhi Wang Zhi-De Deng Luis J. Gomez Elmore Family School of Electrical and Computer Engineering Purdue University WL Indiana USA National Institute of Mental Health Intramural Research Program National Institutes of Health USA

ISBN: (数字)9781946815200

ISBN: (纸本)9798331513979

Transcranial Magnetic Stimulation (TMS) is a non-invasive brain stimulation technique used for the treatment of depression, as well as various neurological and psychiatric disorders. There has been ongoing interest in developing fast E-field solvers that can be easily incorporated into neuronavigation systems to facilitate real-time E-field in targeted brain regions, optimizing dosing parameters.

关键词： Transcranial magnetic stimulation Depression Real-time systems Brain stimulation Optimization

来源：评论

学校读者我要写书评

暂无评论

Towards the Improvement of the Cognitive Process of the Synthesized Speech of Mathematical Expression in MathML: An Eye-Tracking 2

Towards the Improvement of the Cognitive Process of the Synt...

引用

2nd International Conference on Interactive Media, Smart Systems and Emerging Technologies, IMET 2022

作者： Souza, Adriana Freitas, Diamantino Bahia Federal Institute of Education Science and Technology Bahia Brazil Doctoral Program in Digital Media Faculty of Engineering of the University of Porto Porto Portugal Faculty of Engineering of the University of Porto Speech Processing Laboratory Electrical and Computer Engineering Department Porto Portugal

ISBN: (数字)9781665470162

ISBN: (纸本)9781665470162

Being one essential part of the solutions we are developing to provide accessibility for blind persons, synthesized speech of mathematical content, although having evolved in naturalness in recent years, still keeps a characteristic of generating very long outputs, which makes it difficult for a person who uses a screen reader to memorize. Eye-gaze tracking has been useful for studying cognitive processes and student's effort in performing tasks therefore it seems attractive as a means to understand how people analyze, process, and break down extensive math expressions into smaller ones. This understanding can help to build a similar model of auditory rendering of MathML capable of avoiding cognitive overload. Therefore, we carried out an experiment to do a gaze-based analysis of reading and memorizing mathematical expressions. Our sample consisted of fourteen participants, and seven mathematical expressions were used in the experiment. The results showed that the computed complexity of the expressions and the average duration of eye-gaze fixations have a strong and significant relationship. Therefore, we conclude that the complexity variable must always be considered in the speech synthesis process of mathematical expressions. However, when considering why participants requested the repetition of an expression with not larger complexity during the memorization process, a tentative interpretation based on the number of MathML tree nodes and the computed complexity of the expressions was not enough to explain it, indicating that other factors need to be investigated. Although some answers to the research questions were not yet conclusive, we consider this research relevant because it points to directions for the work in order to minimize the cognitive processing and mental load of the synthesized speech rendering of mathematical expressions. © 2022 IEEE.

关键词： Eye tracking

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Deep Learning Models for Detection of Knee Osteoarthritis Disease through Mobile Apps 8

A Comparative Analysis of Deep Learning Models for Detection...

引用

8th International Conference on Wireless and Telematics, ICWT 2022

作者： Kardawi, Muhammad Yusuf Muchtar, Kahlil Nurdin, Yudha Maulina, Novi Muchallil, Sayed Universitas Syiah Kuala Department of Electrical and Computer Engineering Aceh Indonesia Aceh Indonesia Postgraduate Program School of Medicine Universitas Syiah Kuala Microbiology Department Universitas Syiah Kuala Aceh Indonesia

ISBN: (数字)9781665451222

ISBN: (纸本)9781665451222

Osteoarthritis is a progressive degenerative joint disease characterized by damage to joint cartilage and structural joint diarthrodial. As many as 80% of osteoarthritis patients will experience movement limitations, and 25% cannot carry out daily activities. At present, the stages of osteoarthritis diagnosis are through visual observations of the narrowing of joint and osteophyte gaps contained in X-ray images through the observation of radiology and assessing the severity of disorders. This study proposes a CAD (computer-Aided Design) support system to detect knee osteoarthritis based on X-ray images with deep learning (DL) techniques. In the design of the CAD-related architecture evaluation, including ResNet, VggNet, EfficientNet, and InceptionV3 in building a predictive model using the Convolutional Neural Network (CNN) based on Mobile Application and Cloud Computing to be able to classify normal knee and knee osteoarthritis with the percentage of testing accuracy each architecture is 93.17% on ResNet, 95.26% on VggNet, 94.64% in InceptionV3 and 96.59% on EfficientNet. The experimental results can be found through our project channel: https://***/EXj-7Dv6xGM © 2022 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation

arXiv

引用

arXiv 2024年

作者： Oh, Yeongtak Lee, Jonghyun Choi, Jooyoung Jung, Dahuin Hwang, Uiwon Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of School of Computer Science and Engineering Soongsil University Korea Republic of Division of Digital Healthcare Yonsei University Korea Republic of

Test-time adaptation (TTA) addresses the unforeseen distribution shifts occurring during test time. In TTA, performance, memory consumption, and time consumption are crucial considerations. A recent diffusion-based TTA approach for restoring corrupted images involves image-level updates. However, using pixel space diffusion significantly increases resource requirements compared to conventional model updating TTA approaches, revealing limitations as a TTA method. To address this, we propose a novel TTA method that leverages an image editing model based on a latent diffusion model (LDM) and fine-tunes it using our newly introduced corruption modeling scheme. This scheme enhances the robustness of the diffusion model against distribution shifts by creating (clean, corrupted) image pairs and fine-tuning the model to edit corrupted images into clean ones. Moreover, we introduce a distilled variant to accelerate the model for corruption editing using only 4 network function evaluations (NFEs). We extensively validated our method across various architectures and datasets including image and video domains. Our model achieves the best performance with a 100 times faster runtime than that of a diffusion-based baseline. Furthermore, it is three times faster than the previous model updating TTA method that utilizes data augmentation, making an image-level updating approach more feasible. ∗ © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Minimization of fanout in switching networks 15

Minimization of fanout in switching networks

引用

15th Annual Symposium on Switching and Automata Theory, SWAT 1974

作者： Hayes, John P. Department of Electrical Engineering and Computer Science Program University of Southern California Los AngelesCA90007 United States

The problem of finding the minimum amount of fanout needed to realize a switching function f is investigated. Fanout-free functions are defined, and necessary and sufficient conditions for a function to be fanout-free are derived. A measure '(f) called the input fanout index, is introduced which represents the minimum number of input variables that fan out in any realization of f. I is shown that t(f), can be determined from the prime implicants and implicates of f. Another measure of fanout u(f), which is the minimum numbe r of signal lines that must fan out in any network realizing f is defined, and some of the properties of u(f) are examined. © 2021 IEEE. All rights reserved.

关键词： Switching networks

来源：评论

学校读者我要写书评

暂无评论

Development of Support Vector Machine Method to Forecast Dengue-Endemic Areas

Development of Support Vector Machine Method to Forecast Den...

引用

2023 International Conference on Informatics, Multimedia, Cyber and Information Systems, ICIMCIS 2023

作者： Meileni, Hetty Ermatita Abdiansah Husni, Nyayu Latifah Desmawati Doctoral Program in Engineering Science Palembang30139 Indonesia Universitas Sriwijaya Faculty of Computer Science Palembang30139 Indonesia Palembang30139 Indonesia Politeknik Negeri Sriwijaya Palembang Department of Electrical Engineering Palembang30139 Indonesia Universitas Pembangunan Nasional Veteran Jakarta Faculty of Health Science Jakarta12450 Indonesia

ISBN: (纸本)9798350307306

The cases of dengue hemorrhagic fever (DHF) in Indonesia have increased significantly since 2020. Data shown by the Central Statistics Agency, for example, in South Sumatra Province, there were 6,348 cases of DHF during that period. To suppress the case number of DHF, forecasting areas that are susceptible to this disease is urgently needed. This research aims to develop a machine learning method Support Vector Machine (SVM). The data used include area data, population density data, weather data, and dengue fever case data. The development of the SVM model is carried out using a machine learning method that combines regional data, population density data, weather data, and dengue case data. In the end, using the SVM method can be used to forecast dengue-endemic areas, so that the government and society can take further action to reduce the negative impact of DHF. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

NEW INSIGHTS FOR THE STABILITY-PLASTICITY DILEMMA IN ONLINE CONTINUAL LEARNING 11

NEW INSIGHTS FOR THE STABILITY-PLASTICITY DILEMMA IN ONLINE ...

引用

11th International Conference on Learning Representations, ICLR 2023

作者： Jung, Dahuin Lee, Dongjin Hong, Sunwon Jang, Hyemi Bae, Ho Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of DeepBrain AI Inc. Korea Republic of Department of Cyber Security Ewha Womans University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul08826 Korea Republic of

The aim of continual learning is to learn new tasks continuously (i.e., plasticity) without forgetting previously learned knowledge from old tasks (i.e., stability). In the scenario of online continual learning, wherein data comes strictly in a streaming manner, the plasticity of online continual learning is more vulnerable than offline continual learning because the training signal that can be obtained from a single data point is limited. To overcome the stability-plasticity dilemma in online continual learning, we propose an online continual learning framework named multi-scale feature adaptation network (MuFAN) that utilizes a richer context encoding extracted from different levels of a pre-trained network. Additionally, we introduce a novel structure-wise distillation loss and replace the commonly used batch normalization layer with a newly proposed stability-plasticity normalization module to train MuFAN that simultaneously maintains high plasticity and stability. MuFAN outperforms other state-of-the-art continual learning methods on the SVHN, CIFAR100, miniImageNet, and CORe50 datasets. Extensive experiments and ablation studies validate the significance and scalability of each proposed component: 1) multi-scale feature maps from a pre-trained encoder, 2) the structure-wise distillation loss, and 3) the stability-plasticity normalization module in MuFAN. Code is publicly available at https://***/whitesnowdrop/MuFAN. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Stability

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：