检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wisnu, Dyah A.M.G. Rini, Stefano Zezario, Ryandhimas E. Wang, Hsin-Min Tsao, Yu Taiwan International Graduate Program—Social Network and Human Centered Computing Institute of Information Science Academia Sinica Taipei Taiwan College of Informatics National Chengchi University Taipei Taiwan Institute of Communications Engineering National Yang Ming Chiao Tung University Hsinchu Taiwan Research Center for Information Technology Innovation Academia Sinica Taipei Taiwan Institute of Information Science Academia Sinica Taipei Taiwan

This paper introduces HAAQI-Net, a non-intrusive deep learning-based music audio quality assessment model for hearing aid users. Unlike traditional methods like the Hearing Aid Audio Quality Index (HAAQI) that require intrusive reference signal comparisons, HAAQI-Net offers a more accessible and computationally efficient alternative. By utilizing a Bidirectional Long Short-Term Memory (BLSTM) architecture with attention mechanisms and features extracted from the pre-trained BEATs model, it can predict HAAQI scores directly from music audio clips and hearing loss patterns. Experimental results demonstrate HAAQI-Net’s effectiveness, achieving a Linear Correlation Coefficient (LCC) of 0.9368, a Spearman’s Rank Correlation Coefficient (SRCC) of 0.9486, and a Mean Squared Error (MSE) of 0.0064 and inference time significantly reduces from 62.52 to 2.54 seconds. To address computational overhead, a knowledge distillation strategy was applied, reducing parameters by 75.85% and inference time by 96.46%, while maintaining strong performance (LCC: 0.9071, SRCC: 0.9307, MSE: 0.0091). To expand its capabilities, HAAQI-Net was adapted to predict subjective human scores like the Mean Opinion Score (MOS) through fine-tuning. This adaptation significantly improved prediction accuracy, validated through statistical analysis. Furthermore, the robustness of HAAQI-Net was evaluated under varying Sound Pressure Level (SPL) conditions, revealing optimal performance at a reference SPL of 65 dB, with accuracy gradually decreasing as SPL deviated from this point. The advancements in subjective score prediction, SPL robustness, and computational efficiency position HAAQI-Net as a scalable solution for music audio quality assessment in hearing aid applications, contributing to efficient and accurate models in audio signal processing and hearing aid technology. Copyright © 2024, The Authors. All rights reserved.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

First Demonstration of Defect Elimination for Cryogenic Ge FinFET CMOS Inverter Showing Steep Subthreshold Slope by Using Ge-on-Insulator Structure

First Demonstration of Defect Elimination for Cryogenic Ge F...

引用

2023 International Electron Devices Meeting, IEDM 2023

作者： Yu, X.-R. Hsieh, C.-C. Chuang, M.-H. Chiu, M.-Y. Sun, T.-C. Geng, W.-Z. Chang, W.-H. Shih, Y.-J. Lu, W.-H. Chang, W.-C. Lin, Y.-C. Pai, Y.-C. Lai, C.-Y. Chuang, M.-H. Dei, Y. Yang, C.-Y. Lu, H.-Y. Lin, N.-C. Wu, C.-T. Kao, K.-H. Ma, W.C.-Y. Lu, D.D. Lee, Y.-J. Luo, G.-L. Chiang, M.-H. Maeda, T. Wu, W.-F. Li, Y.-M. Hou, T.-H. National Cheng Kung University Dept of Electrical Engineering Tainan Taiwan National Cheng Kung University Academy of Innovative Semiconductor and Sustainable Manufacturing Tainan Taiwan National Yang Ming Chiao Tung University Inst. of Communications Engineering Hsinchu Taiwan National. Sun Yat-Sen University Dept. of Electrical Engineering Kaohsiung Taiwan National Institute of Advanced Industrial Science and Technology Ibaraki Japan National Yang Ming Chiao Tung University Inst. of Artificial Intelligence Innovation Hsinchu Taiwan Taiwan Semiconductor Research Institute Taiwan Intelligent Computing Industrial Program Ming Wu School of Computing National Cheng Kung University Taiwan Inst. of Biomedical Engineering National Yang Ming Chiao Tung University Hsinchu Taiwan Inst. of Pioneer Semiconductor Innovation National Yang Ming Chiao Tung University Hsinchu Taiwan

ISBN: (纸本)9798350327670

This work presents experimental electrical characteristics and circuit prediction at cryogenic temperatures (down to 10 K) for three different kinds of germanium (Ge)-based FETs with advanced Fin/GAA structures. Among them, the layer transferred Ge-on-Insulator (GeOI) FinFET significantly improves its I-V characteristic during cryogenic measurements, such as a steeper subthreshold swing at 10K and a better Ion. The developed GeOI fabrication method provides an effective way to eliminate the defects originating from misfit dislocations at the Ge/Si substrate during epitaxial growth, which would be treated as the key to device performance enhancement under 10 K. According to the measured IV at 10 K and circuit prediction, GeOI FinFETs with high Ge crystallinity are strong candidates for High-Performance-computing (HPC) applications. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Turing Project: An Open Educational Game to Teach and Learn programming Logic 15

Turing Project: An Open Educational Game to Teach and Learn ...

引用

15th Iberian Conference on Information Systems and Technologies, CISTI 2020

作者： Silva, Josivan P. Silveira, Ismar F. Kamimura, Leonardo Barboza, Alcides T. Post-graduation Program on Electrical Eng. And Computing Mackenzie Presbyterian University São Paulo Brazil Cruzeiro Do sul University São Paulo Department of Informatics Brazil

ISBN: (纸本)9789895465903

The paper proposes an open educational game, for students who have never had contact or who present some difficulty in programming logic. The game, in a playful and fun way, tries to teach the basics of programming logic for the students, allowing the same to acquire knowledge during the gameplay. The game proposed in this paper is called 'Turing Project', a RPG game, in which the student will play the role of the game's hero. By collecting scrolls, items and performing missions, students are expected to learn the concepts related to the logic of programming. A subjective experiment was carried out via a survey with 19 users and the main conclusions points out that students had good experiences during the game, which had positive impact in their learning. © 2020 AISTI.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

CapST: An Enhanced and Lightweight Method for Deepfake Video Classification

arXiv

引用

arXiv 2023年

作者： Ahmad, Wasim Peng, Yan-Tsung Chang, Yuan-Hao Ganfure, Gaddisa Olani Khan, Sarwar Shahzad, Sahibzada Adil Institute of Information Science Academia Sinica Taipei11529 Taiwan Department of Computer Science National Chengchi University Taipei116 Taiwan Social Networks Human-Centered Computing Taiwan International Graduate Program Academia Sinica Taipei11529 Taiwan Dire Dawa University Ethiopia Research Center for Information Technology Innovation Academia Sinica Taipei11529 Taiwan

The proliferation of deepfake videos, synthetic media produced through advanced Artificial Intelligence techniques, has raised significant concerns across various sectors, encompassing realms such as politics, entertainment, and security. In response, this research introduces an innovative and streamlined model designed to classify deepfake videos generated by five distinct encoders adeptly. Our approach not only achieves state-of-the-art performance but also optimizes computational resources. At its core, our solution employs part of a VGG19_bn as a backbone to efficiently extract features, a strategy proven effective in image-related tasks. We integrate a Capsule Network coupled with a Spatial-Temporal attention mechanism to bolster the model’s classification capabilities while conserving resources. This combination captures intricate hierarchies among features, facilitating robust identification of deepfake attributes. Delving into the intricacies of our innovation, we introduce an existing video-level fusion technique that artfully capitalizes on temporal attention mechanisms. This mechanism serves to handle concatenated feature vectors, capitalizing on the intrinsic temporal dependencies embedded within deepfake videos. By aggregating insights across frames, our model gains a holistic comprehension of video content, resulting in more precise predictions. Experimental results on an extensive benchmark dataset of deepfake videos (DFDM) showcase the efficacy of our proposed method. Notably, our approach achieves up to a 4% improvement in accurately categorizing deepfake videos compared to baseline models, all while demanding fewer computational resources. Copyright © 2023, The Authors. All rights reserved.

关键词： Routing algorithms

来源：评论

学校读者我要写书评

暂无评论

A physics informed neural network approach to simulating ice dynamics governed by the shallow ice approximation

arXiv

引用

arXiv 2025年

作者： Chawla, Kapil Holmes, William Department of Mathematics Institute for Scientific Computing and Applied Mathematics Indiana University IN47405 United States Department of Mathematics Cognitive Science Program Indiana University IN47405 United States

In this article we develop a Physics Informed Neural Network (PINN) approach to simulate ice sheet dynamics governed by the Shallow Ice Approximation. This problem takes the form of a time-dependent parabolic obstacle problem [17]. Prior work has used this approach to address the stationary obstacle problem and here we extend it to the time dependent problem [4]. Through comprehensive 1D and 2D simulations, we validate the model’s effectiveness in capturing complex free-boundary conditions. By merging traditional mathematical modeling with cutting-edge deep learning methods, this approach provides a scalable and robust solution for predicting temporal variations in ice thickness. To illustrate this approach in a real world setting, we simulate the dynamics of the Devon Ice Cap, incorporating aerogeophysical data from 2000 [6] and 2018 [22]. © 2025, CC BY.

关键词： Approximation theory

来源：评论

学校读者我要写书评

暂无评论

Masked Vascular Structure Segmentation and Completion in Retinal Images

引用

IEEE Transactions on Medical Imaging 2025年第6期44卷 2492-2503页

作者： Yi Zhou Thiara Sana Ahmed Meng Wang Eric A. Newman Leopold Schmetterer Huazhu Fu Jun Cheng Bingyao Tan Singapore National Eye Centre Singapore Eye Research Institute 11 Third Hospital Ave Singapore Beth Israel Deaconess Medical Center Harvard Medical School Boston MA USA Department of Neuroscience University of Minnesota Minneapolis MN USA SERI-NTU Advanced Ocular Engineering (STANCE) Program Discovery Tower Singapore Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP) Duke-NUS Medical School 8 College Rd Singapore Institute of Molecular and Clinical Ophthalmology Basel Switzerland School of Chemical and Biomedical Engineering Nanyang Technological University (NTU) 50 Nanyang Ave Singapore Department of Clinical Pharmacology Medical Physics and Biomedical Engineering Medical University of Vienna Vienna Austria Rothschild Foundation Hospital Paris France Technology and Research Institute of High Performance Computing Agency for Science North Tower Singapore Institute for Infocomm Research Agency for Science Technology and Research Connexis Singapore SERI-NTU Advanced Ocular Engineering Program 50 Nanyang Ave Singapore

Early retinal vascular changes in diseases such as diabetic retinopathy often occur at a microscopic level. Accurate evaluation of retinal vascular networks at a micro-level could significantly improve our understanding of angiopathology and potentially aid ophthalmologists in disease assessment and management. Multiple angiogram-related retinal imaging modalities, including fundus, optical coherence tomography angiography, and fluorescence angiography, project continuous, inter-connected retinal microvascular networks into imaging domains. However, extracting the microvascular network, which includes arterioles, venules, and capillaries, is challenging due to the limited contrast and resolution. As a result, the vascular network often appears as fragmented segments. In this paper, we propose a backbone-agnostic Masked Vascular Structure Segmentation and Completion (MaskVSC) method to reconstruct the retinal vascular network. MaskVSC simulates missing sections of blood vessels and uses this simulation to train the model to predict the missing parts and their connections. This approach simulates highly heterogeneous forms of vessel breaks and mitigates the need for massive data labeling. Accordingly, we introduce a connectivity loss function that penalizes interruptions in the vascular network. Our findings show that masking 40% of the segments yields optimal performance in reconstructing the interconnected vascular network. We test our method on three different types of retinal images across five separate datasets. The results demonstrate that MaskVSC outperforms state-of-the-art methods in maintaining vascular network completeness and segmentation accuracy. Furthermore, MaskVSC has been introduced to different segmentation backbones and has successfully improved performance. The code and 2PFM data are available at: https://***/Zhouyi-Zura/MaskVSC .

关键词： Retina Image segmentation Biomedical imaging Imaging Angiography Training Image color analysis Fluorescence Accuracy Optical imaging

来源：评论

学校读者我要写书评

暂无评论

Exploring NLMS and IPNLMS Adaptive Filtering VLSI Hardware Architectures for Robust EEG Signal Artifacts Elimination

Exploring NLMS and IPNLMS Adaptive Filtering VLSI Hardware A...

引用

IEEE International Conference on Electronics, Circuits and Systems (ICECS)

作者： Andrei La Rosa Patrícia Ucker Guilherme Paim Eduardo da Costa Sérgio Almeida Sergio Bampi Graduate Program on Computing Federal University of Pelotas (UFPel) Brazil Graduate Program on Electronic Engineering and Computing Catholic University of Pelotas (UCPel) Brazil Graduate Program on Microelectronics (PGMicro) Federal University of Rio Grande do Sul (UFRGS) Brazil

ISBN: (数字)9781728160443

ISBN: (纸本)9781728160450

The elimination of artifacts is a crucial procedure to extract the full potential of the information into the EEG processing. Embedded systems require low circuit area (cost) and efficient hardware architectures. Adaptive filtering plays a vital role as a single method and hybrid approaches for robustly eliminating the noise of the real-world EEG measures. In this paper, we propose and implement hardware architectures for both NLMS and IPNLMS adaptive filters. We investigate the filtering performance and circuit area, timing, and power dissipation of the hardware architecture proposals. To leverage the power-efficiency, we improve our hardware architectures employing the data-gating circuit design technique, which provided up to 20% power savings. Also, applying an HDL-Simulink co-simulation, we perform the hardware architectures tradeoff comparing the synthesis results and filtering performance. Our investigation demonstrates that IPNLMS reduces the time domain error in 11%, increasing less than 1% of circuit area and about the double of the energy per operation, compared to the NLMS.

关键词： Electroencephalography Hardware Computer architecture Convergence Measurement Clocks Proposals

来源：评论

学校读者我要写书评

暂无评论

Deep learning outperforms classical machine learning methods in pediatric brain tumor classification through mass spectra

引用

Intelligence-Based Medicine 2024年 10卷 100178-100178页

作者： Santos Bezerra, Thais Maria Silva de Deus, Matheus Cavalaro, Felipe Ribeiro, Denise Seidinger, Ana Luiza Cardinalli, Izilda Aparecida de Melo Porcari, Andreia de Souza Queiroz, Luciano Pedrini, Helio Meidanis, Joao Internship Program in Pediatric Oncology Boldrini Children's Center Sao Paulo Brazil School of Medical Sciences University of Campinas Sao Paulo Brazil Institute of Computing University of Campinas Sao Paulo Brazil MS4Life Laboratory of Mass Spectrometry Health Sciences Postgraduate Program Sao Francisco University Sao Paulo Brazil Boldrini Research Center Boldrini Children's Center Sao Paulo Brazil Boldrini Children's Center Sao Paulo Brazil Scylla Informatics Sao Paulo Brazil

Pediatric brain tumors are the most common cause of death among all childhood cancers and surgical resection usually is the first step in disease management. During surgery, it is important to perform safe gross resection of tumors, retaining as much brain tissue as possible. Therefore, appropriate resection margin delineation is extremely relevant. Currently available methods for tissue analysis have limited precision, are time-consuming, and often require multiple invasive procedures. Our main goal is to test whether machine learning techniques are capable of classifying the pediatric brain tissue chemical profile generated by DESI-MSI, which is mainly lipidic, into normal or abnormal tissue and into low- and high-grade malignancy subareas within each sample. Our experiments show that deep learning methods outperform classical machine learning methods in the task of classifying brain tissue from DESI-MSI mass spectra, both in normal versus abnormal tissue, and, for malignant tissues, in low-grade versus high-grade malignancy. Our conclusion are based on the analysis of 34,870 annotated spectra, obtained from the neoplastic and non-neoplastic microanatomical stratification of individual samples from 116 pediatric patients who underwent brain tumor surgical resection at the Boldrini Children's Center between 2000 and 2020. Support Vector Machines, Random, Forests, and Least Absolute Shrinkage and Selection Operator (LASSO) were among the classical machine learning techniques evaluated. © 2024

关键词： LASSO Neural networks Random forests Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Towards Adversarial Robustness for Multi-Mode Data through Metric Learning

引用

Sensors 2023年第13期23卷 6173页

作者： Khan, Sarwar Chen, Jun-Cheng Liao, Wen-Hung Chen, Chu-Song Research Center for Information Technology Innovation Academia Sinica Taipei11529 Taiwan Social Networks Human-Centered Computing Taiwan International Graduate Program Academia Sinica Taipei11529 Taiwan Department of Computer Science National Chengchi University Taipei11605 Taiwan Department of Computer Science and Information Engineering National Taiwan University Taipei106319 Taiwan

Adversarial attacks have become one of the most serious security issues in widely used deep neural networks. Even though real-world datasets usually have large intra-variations or multiple modes, most adversarial defense methods, such as adversarial training, which is currently one of the most effective defense methods, mainly focus on the single-mode setting and thus fail to capture the full data representation to defend against adversarial attacks. To confront this challenge, we propose a novel multi-prototype metric learning regularization for adversarial training which can effectively enhance the defense capability of adversarial training by preventing the latent representation of the adversarial example changing a lot from its clean one. With extensive experiments on CIFAR10, CIFAR100, MNIST, and Tiny ImageNet, the evaluation results show the proposed method improves the performance of different state-of-the-art adversarial training methods without additional computational cost. Furthermore, besides Tiny ImageNet, in the multi-prototype CIFAR10 and CIFAR100 where we reorganize the whole datasets of CIFAR10 and CIFAR100 into two and ten classes, respectively, the proposed method outperforms the state-of-the-art approach by 2.22% and 1.65%, respectively. Furthermore, the proposed multi-prototype method also outperforms its single-prototype version and other commonly used deep metric learning approaches as regularization for adversarial training and thus further demonstrates its effectiveness. © 2023 by the authors.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Fault Tolerant Quantum Error Mitigation

arXiv

引用

arXiv 2023年

作者： Gonzales, Alvin Babu, Anjala M. Liu, Ji Saleem, Zain Byrd, Mark Intelligence Community Postdoctoral Research Fellowship Program Argonne National Laboratory LemontIL United States School of Physics and Applied Physics Southern Illinois University CarbondaleIL United States Mathematics and Computer Science Division Argonne National Laboratory LemontIL United States School of Computing Southern Illinois University CarbondaleIL United States

Typically, fault-tolerant operations and code concatenation are reserved for quantum error correction due to their resource overhead. Here, we show that fault tolerant operations have a large impact on the performance of symmetry based error mitigation techniques. We also demonstrate that similar to results in fault tolerant quantum computing, code concatenation in fault-tolerant quantum error mitigation (FTQEM) can exponentially suppress the errors to arbitrary levels. For a family of circuits, we provide analytical error thresholds for FTQEM with the repetition code. These circuits include a set of quantum circuits that can generate all of reversible classical computing. The post-selection rate in FTQEM can also be increased by correcting some of the outcomes. Our threshold results can also be viewed from the perspective of quantifying the number of gate operations we can delay checking the stabilizers in a concatenated code before errors overwhelm the encoding. The benefits of FTQEM are demonstrated with numerical simulations and hardware demonstrations. Copyright © 2023, The Authors. All rights reserved.

关键词： Quantum computers

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：