检索结果-内蒙古大学图书馆

Snippet Policy Network V2: Knee-Guided Neuroevolution for Multi-Lead ECG Early Classification

IEEE Transactions on Neural Networks and Learning systems 2024年第2期35卷 2167-2181页

作者： Huang, Yu Yen, Gary G. Tseng, Vincent S. National Yang Ming Chiao Tung University Department of Computer Science Hsinchu300 Taiwan Oklahoma State University School of Electrical and Computer Engineering StillwaterOK74078 United States National Chung Hsing University Department of Management Information Systems Taichung402 Taiwan

Early time series classification predicts the class label of a given time series before it is completely observed. In time-critical applications, such as arrhythmia monitoring in ICU, early treatment contributes to the patient's fast recovery, and early warning could even save lives. Hence, in these cases, it is worthy of trading, to some extent, classification accuracy in favor of earlier decisions when the time series data are collected over time. In this article, we propose a novel deep reinforcement learning-based framework, snippet policy network V2 (SPN-V2), for long and varied-length multi-lead electrocardiogram (ECG) early classification. The proposed SNP-V2 contains two main components: snippet representation learning (SRL) and early classification timing learning (ECTL). The SRL is proposed to encode inner-snippet spatial correlations and inter-snippet temporal correlations into the hidden representations of the subsegment (snippet) of the input ECG. ECTL aims to learn a decision agent to classify the time series early and accurately. To optimize the proposed framework, we design a novel knee-guided neuroevolution algorithm (KGNA) to solve cardiovascular diseases' early classification problem, automatically optimizing the proposed SPN-V2 regarding the tradeoff between accuracy and earliness. In addition, we conduct a series of experiments on two real-world ECG datasets. The experimental results show the superiority of the proposed algorithm over the state-of-the-art competing methods. © 2012 IEEE.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Diabetic and Hypertensive Retinopathy Classification from Retinal Images Using Dual Vision Transformer 4

Diabetic and Hypertensive Retinopathy Classification from Re...

引用

4th International Conference on Digital Data Processing, DDP 2024

作者： Koduri, Harsha Ma, Ming Katz School of Science and Health Yeshiva University New York United States Katz School of Science and Health Yeshiva University Department of Graduate Computer Science and Engineering New York United States

ISBN: (纸本)9798331515706

Accurate and automated classification of diabetic and hypertensive retinopathy conditions plays an important role in the diagnosis as it can lead to severe vision loss if not promptly diagnosed and treated. We propose to use a multi-scale method via dual vision transformer to improve the detection and classification of these retinopathy conditions. The method is capable of extracting more holistic global view of semantic feature and learning finer local features. To demonstrate the effectiveness of the method in this work, comparative experiments against the state-of-the-art methods are conducted. Experimental results show that the method in this work achieves better performance in terms of various metrics including accuracy, F1 score, precision, and recall. This work underscores the potential of employing multi-scale dual vision transformer method to enhance the reliability and precision of automated retinal disease classification, offering significant benefits for early diagnosis and patient care. © 2024 IEEE.

关键词： classification diabetic retinopathy hypertensive retinopathy multi-scale retinal images vision transformer

来源：评论

学校读者我要写书评

暂无评论

A Low-Cost Semihosting Approach to Debug DSP Application 17th

A Low-Cost Semihosting Approach to Debug DSP Application

引用

17th International Conference on Communications and Networking, ChinaCom 2022

作者： Huang, Tao Ren, Haoqi Zhang, Zhifeng Tan, Bin Wu, Jun Department of Computer Science Tongji University Shanghai China School of Electronic and Information Engineering Jinggangshan University Ji’an China School of Computer Science Fudan University Shanghai China

ISBN: (纸本)9783031347894

Applications of digital signal processor (DSP) involve large amounts of data processing. In order to be able to improve the speed of DSP application development, it is necessary to be able to implement debugging functions that can support File I/O at a small cost of modification. This paper proposes a complete set of implementation methods of the semihosting debugging function on a DSP without an operating system, including hardware support and software algorithms. On hardware, this article adds only one self-trapping instruction to support semihosting debugging. In software, this article is based on the GDB File I/O extension protocol to design and implement the debugging agent software and C language library underlying file operations. And after optimizing library files and application source code, the I/O speed of the architecture can meet users’ debugging needs. This solution can realize the semihosting debugging function for DSP chip at a low cost and has good performance. Therefore, provides more powerful debugging functions for DSP application development. In assembly level debugging, observing the execution of each assembly instruction can also verify the instruction execution of the chip itself. The addition of file operation can greatly validate the chip with larger data. © 2023, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： Digital signal processing

来源：评论

学校读者我要写书评

暂无评论

A parametric survey on polyphonic sound event detection and localization

引用

Multimedia Tools and Applications 2024年 1-38页

作者： Mohmmad, Sallauddin Sanampudi, Suresh Kumar JNTU Hyderabad India School of Computer Science and Artificial Intelligence SR University TS Warangal India Department of Information Technology JNTUH University College of Engineering Nachupally Kondagattu TS Jagtial India

With the rapid growth in technology, everything will become automatic. Automatic detection is needed in health, security, smart homes, smart cities, and various environments. Sound is an essential function of automatic detection, which can be created in different environmental situations. To handle this, various sound detection devices are incorporated that run alone or with a combination of video. This auditory detection system is becoming more emerging in research today. This study is a survey on sound event detection and localization-related models. This survey research included details about different datasets, feature extractions, and classification methods that are implemented in sound-related research. Distinct and typical environmental situations comprise multiple sound sources that generate different sounds and noise. Our research compares distinct algorithms to detect sound events and localization, such as HMM, CNN, SVM, Random Forest, NMF, CRNN, VGG16, etc. Sound-related feature extraction methods, such as MFCC, Mel-Spectrogram, Mel band elements, ZCR, and wavelet features, and their importance in sound classification are also explained with a comparative analysis approach. The sections end with challenges with the existing approaches in different environments and feature extraction combinations. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

GROOT: LEARNING TO FOLLOW INSTRUCTIONS BY WATCHING GAMEPLAY VIDEOS 12

GROOT: LEARNING TO FOLLOW INSTRUCTIONS BY WATCHING GAMEPLAY ...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Cai, Shaofei Zhang, Bowei Wang, Zihao Ma, Xiaojian Liu, Anji Liang, Yitao Institute for Artificial Intelligence Peking University China School of Intelligence Science and Technology Peking University China School of Electronics Engineering and Computer Science Peking University China Computer Science Department University of California Los Angeles United States China

We study the problem of building an agent that can follow open-ended instructions in open-world environments. We propose to follow reference videos as instructions, which offer expressive goal specifications while eliminating the need for expensive text-gameplay annotations. We implement our agent GROOT in a simple yet effective encoder-decoder architecture based on causal transformers. We evaluate GROOT against open-world counterparts and human players on a proposed Minecraft SkillForge benchmark. The Elo ratings clearly show that GROOT is closing the human-machine gap as well as exhibiting a 70% winning rate over the best generalist agent baseline. Qualitative analysis of the induced goal space further demonstrates some interesting emergent properties, including the goal composition and complex gameplay behavior synthesis. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multiband Slotted Crescent-shaped Patch Antenna for K-band Satellite and mmWave Communications

引用

IEIE Transactions on Smart Processing and Computing 2022年第3期11卷 213-221页

作者： Hossain, Md Najmul Paul, Liton Chandra Rahim, Md Abdur Shin, Jungpil Department of Electrical Electronic and Communication Engineering Faculty of Engineering and Technology Pabna University of Science and Technology Pabna6600 Bangladesh Department of Computer Science and Engineering Faculty of Engineering and Technology Pabna University of Science and Technology Pabna6600 Bangladesh School of Computer Science and Engineering The University of Aizu Aizuwakamatsu Fukushima965-8580 Japan

In this paper, a multiband miniaturized crescent-shaped patch antenna with circular slots is presented for ultra-wideband applications. The proposed antenna is constructed on a Flame Retardant 4 (FR-4) dielectric substrate. Next-generation wireless communication systems, such as beyond 5G (B5G) and 6G mobile communication systems require a wider bandwidth and higher data rates. The overall size of the proposed antenna is 11.4 mm3 operating in the frequency range of 17.73 GHz to 26.04 GHz and 29.6 GHz to 31.02 GHz for satellite communications and radar systems, respectively, as well as 35.40 GHz to 38.65 GHz for future mmWave B5G mobile applications. These frequency ranges usually fall into the K-band and mmWave frequency spectrums. Simulation results demonstrate that the peak gain of the proposed antenna is 4.68 dB, which is obtained at 25 GHz. © 2022 The Institute of Electronics and Information Engineers.

关键词： Microstrip antennas

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Powered Edge Analytics for IoT-based Sensor Networks 2

Deep Learning-Powered Edge Analytics for IoT-based Sensor Ne...

引用

2nd International Conference on Machine Learning and Autonomous systems, ICMLAS 2025

作者： Karthikeyan, M. Saravana Subramani, Ponnusamy Selvam, M. Panneer Sworna Kokila, M.L. Vijayalakshmi, R. Sundaravadivel, P. Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Avadi Department of Computer Science and Engineering Chennai India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Physics Chennai India Sona College of Technology Department of Electrical and Electronics Engineering Salem India School of Computing SRM Institute of Science and Technology Kattankulathur Faculty of Engineering and Technology Department of Computing Technologies Chennai India Velammal College of Engineering and Technology Department of Computer Science and Engineering Madurai India Saveetha Engineering College Department of Artificial Intelligence and Machine Learning Chennai India

ISBN: (纸本)9798331505745

The rapid expansion of loT-based sensor networks has necessitated the development of efficient edge analytics frameworks to process vast amounts of data in real time while minimizing computational overhead. Deep learning-powered edge computing has emerged as a transformative approach, enhancing data processing capabilities at the network edge while addressing constraints related to latency, bandwidth, and energy consumption. This study introduces a novel algorithm, Lightweight AI Model Compression for Energy-Efficient IoT Edge Devices (LiteEdgeAI), designed to optimize deep learning models for resource-constrained edge environments. The proposed method employs quantization, pruning, and knowledge distillation techniques to achieve significant reductions in model size and computational complexity while maintaining high inference accuracy. To validate the efficiency of LiteEdgeAI, a comparative simulation analysis was conducted against existing state-of-the-art algorithms using key performance metrics such prediction accuracy, energy efficiency and computational complexity. Experimental results demonstrated that LiteEdgeAI outperformed baseline models. The findings highlight the potential of LiteEdgeAI in enabling scalable, low-power, and real-time edge intelligence for IoT-based sensor networks. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Analysis of Bitcoin Prediction using Transparent and Opaque Machine Learning Algorithms 15

Analysis of Bitcoin Prediction using Transparent and Opaque ...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Singh, Kavita R. Jamar, Anushri V. Khadgi, Pracheta P. Kapse, Smita Kumar, Ajay Department of Computer Technology Yeshwantrao Chavan College of Engineering Maharashtra Nagpur India School of Big Data and Computer Science Guizhou Province Guiyang City China

ISBN: (纸本)9798331300579

The unpredictable nature of cryptocurrency markets, particularly Bitcoin, has attracted significant attention from investors, researchers, and financial institutions seeking to understand and predict price movements. This study aims to analyse the effectiveness of both transparent and opaque machine learning algorithms in predicting Bitcoin prices. Transparent algorithms, such as linear regression and decision trees, provide clear insights into the factors influencing predictions. On the other hand, opaque algorithms, like ensemble methods, offer complex, non-linear models that may capture intricate patterns in the data. The main analysis is to lower the mean absolute error and enhance the r2 square, both key indicators of forecast accuracy. The study presents that linear regression and random forest outperformed others, achieving the smallest error and highest variance explanation, suggesting their effectiveness in Bitcoin price prediction with and without outliers respectively. In contrast, the support vector machine underperforms compared to all other machine learning algorithms under consideration. This paper sheds light on the most capable machine-learning techniques for Bitcoin forecasting, aiding trader investment choices and guiding academic study in the field of prediction. © Grenze Scientific Society, 2024.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Towards Temporal Dependency Identification based on Multivariate Time Series IIoT Data 14

Towards Temporal Dependency Identification based on Multivar...

引用

14th International Conference on Information and Communication Technology Convergence, ICTC 2023

作者： Isah, Abubakar Shin, Hyeju Hassan, Sani Ahmad Oh, Seugmin Kim, Jinsul Chonnam National University Department of Ict Convergence System Engineering Gwangju Korea Republic of Ahmadu Bello University Department of Computer Science Zaria Nigeria

ISBN: (纸本)9798350313277

In multivariate time series data, the analysis of temporal dependencies is crucial for comprehending complex relationships, forecasting trends, and streamlining diverse applications. In this study, we proposed a method for improving multivariate time series data using temporal dependence modeling. Our approach is based on utilizing cutting-edge machine learning methods, particularly Long Short-Term Memory (LSTM) networks, that are well known for their capacity to record complicated temporal correlations. We want to reveal hidden patterns and dependencies within multivariate time series data by employing LSTM networks. We use benchmark metrics Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE), and Mean Squared Error (MSE) to assess the effectiveness of our strategy. We run comprehensive tests on various datasets to demonstrate the accuracy and robustness of our model. © 2023 IEEE.

关键词： LSTM multivariate temporal dependency

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Enhanced Detection and Characterization of Pulmonary Nodules 4

Deep Learning for Enhanced Detection and Characterization of...

引用

4th International Conference on Intelligent Technologies, CONIT 2024

作者： Lavanya, G. Muthulakshmi, M. Latha, Manohar Keerthinathan, A. Vishal Krishh, P. Sulthana, Sheba Meghana, Meka Kavya Uma Amrita School of Computing Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Chennai India Amrita School of Computing Department of Electronics and Communication Engineering Amrita Vishwa Vidyapeetham Chennai India Rajalakshmi Engineering College Department of Electronics and Communication Engineering India

ISBN: (纸本)9798350349900

This study aims to develop, an effective customized Convolutional Neural Network model that can accurately identify areas on CT scans that have lung nodules and those that do not. The study utilizes two datasets, the LUNA-16 dataset for training and the IQ-OTH/NCCD dataset for testing. Additionally, we employ a U-Net model for precise lung nodule segmentation. This segmentation approach aimed to achieve a comprehensive characterization of nodules based on their spatial characteristics. The testing phase involves evaluating the performance of our trained models on the IQ-OTH/NCCD dataset, particularly focusing on localizing nodules in malignant cases. Overall, the research highlights the effectiveness of deep learning in early detection and diagnosis of lung cancer. This work contributes to improving the accuracy and reliability of automated pulmonary nodule detection systems, with a focus on identifying nodules in patients with suspected malignancies. © 2024 IEEE.

关键词： Lung cancer

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：