检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Weiss, Jonah Obrien Alves, Tiago Kundu, Sandip Dept. of Electrical and Computer Engineering University of Massachusetts Amherst AmherstMA United States Dept. of Informatics and Computer Science State University of Rio de Janeiro - UERJ Rio de Janeiro Brazil

The prevalence and success of Deep Neural Network (DNN) applications in recent years have motivated research on DNN compression, such as pruning and quantization. These techniques accelerate model inference, reduce power consumption, and reduce the size and complexity of the hardware necessary to run DNNs, all with little to no loss in accuracy. However, since DNNs are vulnerable to adversarial inputs, it is important to consider the relationship between compression and adversarial robustness. In this work, we investigate the adversarial robustness of models produced by several irregular pruning schemes and by 8-bit quantization. Additionally, while conventional pruning removes the least important parameters in a DNN, we investigate the effect of an unconventional pruning method: removing the most important model parameters based on the gradient on adversarial inputs. We call this method Greedy Adversarial Pruning (GAP) and we find that this pruning method results in models that are resistant to transfer attacks from their uncompressed counterparts. Code is available at [1]. © 2022, 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

Visual Field Prediction for Fundus Image with Generative AI

Visual Field Prediction for Fundus Image with Generative AI

引用

International Conference on Ubiquitous Information Management and Communication (IMCOM)

作者： Honggu Kang Seonghyeon Ko Juchan Kim Duc-Tai Le Junghyun Bum Jongchul Han Hyunseung Choo Dept. of AI Systems Engineering Sungkyunkwan University Suwon South Korea College of Computing and Informatics Sungkyunkwan University Suwon South Korea Sungkyun AI Research Institute Sungkyunkwan University Suwon South Korea Dept. of Ophthalmology Samsung Medical Center Seoul South Korea Dept. of Electrical and Computer Engineering Sungkyunkwan University Suwon South Korea

Accurate diagnosis of glaucoma is crucial due to its high risk of blindness, but the long examination time often undermines the reliability of results by examinee's subjective factors. We propose a method to reduce examination time by generating static perimetry results with Conventional Fundus Images (CFIs) utilizing the CFI2GM technique, which leverages multimodal data. Based on data from 3,306 glaucoma patients at Samsung Medical Center in Seoul, we conducted ophthalmic image translation utilizing the Pix2Pix model. Our method, combining cGAN, L1, and SSIM loss, achieved MSE 57.9886 and PSNR 30.6057 dB. Furthermore, we received positive feedback from ophthalmologist regarding the high practical applicability of the images generated by our method. This indicates that CFI2GM can enhance the reliability of glaucoma examination results as well as reduce testing time.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Graph Neural Networks for IoT Data Aggregation Scheduling

Graph Neural Networks for IoT Data Aggregation Scheduling

引用

IEEE Symposium on Network Operations and Management

作者： Van-Vi Vo Syed M. Raza Duc-Tai Le Moonseong Kim Hyunseung Choo Convergence Research Institute Sungkyunkwan University Suwon South Korea Dept. of Electrical and Computer Engineering Sungkyunkwan University Suwon South Korea College of Computing and Informatics Sungkyunkwan University Suwon South Korea Dept. of IT Convergence Software Seoul Theological University Bucheon South Korea

ISBN: (数字)9798350327939

ISBN: (纸本)9798350327946

Data aggregation is an important approach in IoT sensor networks since it reduces data transfer while also preserving energy and bandwidth. This research investigates the challenge of time-efficient data aggregation in wireless sensor networks, which is critical in military, civilian, and industrial applications. Effective data aggregation algorithm design and optimization are required for quick and interference-free data collection. Machine learning has received attention for outperforming classical heuristic techniques. The research presents the first Graph Neural Network (GNN) model for data aggregation in IoT sensor networks, which incorporates Graph Attention Networks (GATs) and fully connected layers. The GNN-based model learns network topology and node attributes, creating node embeddings and correcting sensor node transmitting time slots. With a centralized training procedure and adapted execution for network size change, the proposed approach achieves satisfactory performance compared to the heuristic algorithm.

关键词： Training Adaptation models Wireless sensor networks Machine learning algorithms Job shop scheduling Network topology Machine learning

来源：评论

学校读者我要写书评

暂无评论

Non-Invasive Device for Hearing Rehabilitation

Non-Invasive Device for Hearing Rehabilitation

引用

E-Health and Bioengineering Conference (EHB)

作者： Daniela Goldan Mihai Aron Teofil Ilie Ursache Gladiola Petroiu Cristian Rotariu Dept. of Biomedical Sciences Faculty of Medical Bioengineering Grigore T. Popa University of Medicine and Pharmacy of Iasi Iași Romania Dept. of Automatic Control and Applied Informatics Faculty of Automatic Control and Computer Engineering Gheorghe Asachi Technical University Iași Romania

ISBN: (数字)9798331532147

ISBN: (纸本)9798331532154

Hearing loss impacts over 5% of the world's population, a percent rising due to aging populations, noise exposure, and chronic health conditions. To address this, we developed a non-invasive hearing device designed for individuals with hearing challenges that limit the use of hearing aids. The proposed device integrates a set of Bluetooth headphones with a smartphone, that process the ambient sound in real time through a software application and delivers to the hear impaired patient via bone conduction, bypassing traditional auditory pathways. Laboratory tests confirmed that the proposed device effectively replicates auditory perception. The developed device offers an accessible and user-friendly alternative to patients who cannot use traditional hearing aids.

关键词： Irrigation Noise Auditory system Ear Receivers Hearing aids Propagation losses Real-time systems Software Testing

来源：评论

学校读者我要写书评

暂无评论

Advanced Learning Schemes for Metaverse Applications in B5G/6G Networks

Advanced Learning Schemes for Metaverse Applications in B5G/...

引用

Metaverse Computing, Networking and Applications (MetaCom), IEEE International Conference on

作者： Sherief Hashima Mostafa M. Fouda Kohei Hatano Eiji Takimoto Computational Learning Theory Team RIKEN-AIP Fukuoka Japan Engineering Dept Egyptian Atomic Energy Authority Cairo Egypt Department of Electrical and Computer Engineering Idaho State University Pocatello ID USA Department of Informatics Kyushu University Fukuoka Japan

The metaverse, a virtual world where various users can jointly/independently interact with digital objects seamlessly, is expected to be a key application area in beyond fifth generation (B5G)/ six-generation (6G) networks. However, the metaverse design poses several challenges, such as efficient usage of limited resources and the provision of high-quality user experiences. In this paper, we recommend advanced machine learning techniques such as multi-armed bandits (MABs), federated learning (FL), meta-learning, etc., to address these challenges. We illustrate the evaluability of these techniques with two specific use cases: resource allocation for virtual reality (VR) applications and personalized content delivery in the metaverse.

关键词：

来源：评论

学校读者我要写书评

暂无评论

pylspack: Parallel algorithms and data structures for sketching, column subset selection, regression and leverage scores

arXiv

引用

arXiv 2022年

作者： Sobczyk, Aleksandros Gallopoulos, Efstratios IBM Research Europe ETH Zürich Zürich Switzerland HPCLAB Computer Engineering and Informatics Dept. University of Patras Patras Greece

MSC Codes 65F50, 65F08, 65F20, 68W10, 68W20, 68P05We present parallel algorithms and data structures for three fundamental operations in Numerical Linear Algebra: (i) Gaussian and CountSketch random projections and their combination, (ii) computation of the Gram matrix and (iii) computation of the squared row norms of the product of two matrices, with a special focus on "tall-and-skinny" matrices, which arise in many applications. We provide a detailed analysis of the ubiquitous CountSketch transform and its combination with Gaussian random projections, accounting for memory requirements, computational complexity and workload balancing. We also demonstrate how these results can be applied to column subset selection, least squares regression and leverage scores computation. These tools have been implemented in pylspack, a publicly available Python package, whose core is written in C++ and parallelized with OpenMP, and which is compatible with standard matrix data structures of SciPy and NumPy. Extensive numerical experiments indicate that the proposed algorithms scale well and significantly outperform existing libraries for tall-and-skinny matrices. © 2022, CC BY-NC-ND.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Evaluating BERT-based Pre-training Language Models for Detecting Misinformation

arXiv

引用

arXiv 2022年

作者： Anggrainingsih, Rini Hassan, Ghulam Mubashar Datta, Amitava Informatics Dept Faculty of Mathematics and Natural Sciences Sebelas Maret University Indonesia Department of Computer Science and Software Engineering University of Western Australia Australia

The rapid growth of internet technology and social media platforms has led to rapid changes in communication by providing many conveniences to users. Despite the comforts provided, it is challenging to control the quality of information due to the lack of supervision over all the information posted on online media. For instance, there are many problems in society because of the numerous rumours and the quick spread of fake news through online media. Manual checking is almost impossible given the vast number of posts made on online media and how quickly they spread. Therefore, there is a need for automated rumour detection technique to limit the adverse effects of spreading misinformation. Previous studies mainly focused on finding and extracting the significant features of text data and used standard supervised learning techniques or deep learning approaches to identify misinformation. However, extracting features is timeconsuming and not a highly effective process due to the nonavailability of some features, affecting the effectiveness and performance of the detection models. Therefore, this study proposes the BERT-based pre-trained language models to encode text data into vectors and utilise neural network models to classify these vectors to detect misinformation. Furthermore, the performance of different language models (LM) containing different number of trainable parameters were compared, including RoBERTa, BERT, and DistillBERT. The proposed technique is tested on different twitter and fake news datasets to represent short and long text data respectively. The study also proposes a standardised distribution ratio for taining, validation and testing of the model. The result of the proposed technique has been compared with the state-of-the-art techniques on the same datasets. The results show that the proposed technique performs better than the state-of-the-art techniques. Moreover, the choice of LM either with significantly large or small number of parameters do

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Automatic Parkinson's tremor assessment with data analysis from daily activities 6

Automatic Parkinson's tremor assessment with data analysis f...

引用

6th South-East Europe Design Automation, computer engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2021

作者： Fiska, Vasiliki Gakilazou, Vasiliki Katertsidis, Nikolaos Tzallas, Alexandros T. Giannakeas, Nikolaos Tsipouras, Markos G. Dept. of Electrical and Computer Engineering Univ. of Western Macedonia Kozani Greece Dept. of Informatics and Telecommunications Univ. of Ioannina Arta Greece Science and Technology Park of Epirus UNIVEYE IKE Ioannina Greece

ISBN: (纸本)9781665427425

A new approach for the assessment of motor symptoms caused by Parkinson's disease (PD), based on data recorded using a smart mobile phone, is presented in this manuscript. Data were obtained from the online platform ***, and were analyzed based on machine learning techniques to produce a comprehensive physician report, presenting motor symptoms in everyday life. The idea of this study is to equip PD patients with a smartphone, which will monitor signals in real time for a period of time in order to enhance the medical diagnosis routine of a physician, providing information about the overall picture of the patient's motor condition, resulting in the provision of individualized treatment to the patient. © 2021 IEEE

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

Cardiac Condition Detection Model Using Alexnet and Time-Frequency Representation of ECG Signals

Cardiac Condition Detection Model Using Alexnet and Time-Fre...

引用

E-Health and Bioengineering Conference (EHB)

作者： Mihai Aron Teofil Ilie Ursache Gladiola Petroiu Cristian Rotariu Dept. of Biomedical Sciences Grigore T. Popa University of Medicine and Pharmacy of Iasi Faculty of Medical Bioengineering Iaşi Romania Dept. of Automatic Control and Applied Informatics Gheorghe Asachi Technical University Faculty of Automatic Control and Computer Engineering Iaşi Romania

ISBN: (数字)9798331532147

ISBN: (纸本)9798331532154

Cardiac conditions are one of the leading causes of death worldwide. To address this, we designed a model that can detect cardiac conditions from a patient's electrocardiogram (ECG). This will convert the 1D ECG signal into a 2D scalogram using wavelet transforms with Morlet wavelets. With the help of a network based on the AlexNet convolutional neural network, it distinguishes between a normal ECG signal and one indicative of arrhythmia or congestive heart failure. The resulting model performs well, achieving a classification accuracy of 92.20% and providing a robust approach for detecting cardiac conditions.

关键词： Wavelet transforms Training Time-frequency analysis Accuracy Biological system modeling Electrocardiography Predictive models Congestive heart failure Convolutional neural networks Biomedical engineering

来源：评论

学校读者我要写书评

暂无评论

Load Shedding Automation System on Smart Meter Using Fuzzy Logic Method

Load Shedding Automation System on Smart Meter Using Fuzzy L...

引用

2021 International Conference on Advanced Mechatronics, Intelligent Manufacture and Industrial Automation, ICAMIMIA 2021

作者： Farid, Imam Wahyudi Priananda, Ciptian Weried Yuhansyah, Divan Andika Arifiani, Siska Hariadi, Ridho Rahman Priyadi, Ardyono Purnomo, Mauridhi Hery Institut Teknologi Sepuluh Nopember Dept. Electrical Automation Engineering Dept. Electrical Engineering Surabaya Indonesia Institut Teknologi Sepuluh Nopember Dept. Electrical Automation Engineering Surabaya Indonesia Institut Teknologi Sepuluh Nopember Dept. Informatics Engineering Surabaya Indonesia Institut Teknologi Sepuluh Nopember Dept. Electrical Engineering Surabaya Indonesia Institut Teknologi Sepuluh Nopember Dept. Computer Engineering Surabaya Indonesia

ISBN: (纸本)9781665499699

The need for electrical energy for the household sector has become a basic need. However, there are still people who complain about expensive electricity bills. In addition, the use of electricity is not recommended to exceed the power capacity which can cause a short circuit. Therefore, a tool is needed, namely a Smart Meter that can monitor the use of electrical energy and can perform load shedding automatically to reduce the occurrence of short circuits. The load shedding automation system uses one of the methods of artificial intelligence technology, namely by using Mamdani's Fuzzy Logic so that it can recognize patterns, classify/identify, predict, and optimize the use of electrical energy. Load shedding is carried out based on priority loads and non-priority loads. From testing the load shedding automation system which was carried out 5x with 6x load combination trials, it resulted that in the 1st to 4th experiments, Relay_Hidup3 output was obtained. To carry out the load shedding process, the average delay time is 2,302 seconds © 2021 IEEE.

关键词： Fuzzy logic

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：