检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Tang, Jiankai Liu, Jiacheng Tong, Renling Zhu, Kai Li, Zhe Yi, Xin Xing, Junliang Shi, Yuanchun Wang, Yuntao Key Laboratory of Pervasive Computing Ministry of Education Department of Computer Science and Technology Tsinghua University Ant Group China Key Laboratory of Pervasive Computing Ministry of Education Department of Computer Science and Technology Tsinghua University China Ant Group China Key Laboratory of Pervasive Computing Ministry of Education Department of Computer Science and Technology Tsinghua University China Intelligent Computing Application Laboratory of Qinghai Province Qinghai University China

Photoplethysmography (PPG) Sensors, widely deployed in smartwatches, offer a simple and non-invasive authentication approach for daily use. However, PPG authentication faces reliability issues due to motion artifacts from physical activity and physiological variability over time. To address these challenges, we propose MTL-RAPID, an efficient and reliable PPG authentication model, that employs a multitask joint training strategy, simultaneously assessing signal quality and verifying user identity. The joint optimization of these two tasks in MTL-RAPID results in a structure that outperforms models trained on individual tasks separately, achieving stronger performance with fewer parameters. In our comprehensive user studies regarding motion artifacts (N = 30), time variations (N = 32), and user preferences (N = 16), MTL-RAPID achieves a best AUC of 99.2% and an EER of 3.5%, outperforming existing baselines. We open-source our PPG authentication dataset along with the MTL-RAPID model to facilitate future research on GitHub. © 2025, CC BY.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

SpeechCraft: A Fine-Grained Expressive Speech Dataset with Natural Language Description 24

SpeechCraft: A Fine-Grained Expressive Speech Dataset with N...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Jin, Zeyu Jia, Jia Wang, Qixin Li, Kehan Zhou, Shuoyi Zhou, Songtao Qin, Xiaoyu Wu, Zhiyong Department of Computer Science and Technology Tsinghua University Beijing China BNRist Tsinghua University Key Laboratory of Pervasive Computing Ministry of Education Beijing China Shenzhen International Graduate School Tsinghua University Shenzhen China

ISBN: (纸本)9798400706868

Speech-language multi-modal learning presents a significant challenge due to the fine nuanced information inherent in speech styles. Therefore, a large-scale dataset providing elaborate comprehension of speech style is urgently needed to facilitate insightful interplay between speech audio and natural language. However, constructing such datasets presents a major trade-off between large-scale data collection and high-quality annotation. To tackle this challenge, we propose an automatic speech annotation system for expressiveness interpretation that annotates in-the-wild speech clips with expressive and vivid human language descriptions. Initially, speech audios are processed by a series of expert classifiers and captioning models to capture diverse speech characteristics, followed by a fine-tuned LlaMA for customized annotation generation. Unlike previous tag/templet-based annotation frameworks with limited information and diversity, our system provides in-depth understandings of speech style through tailored natural language descriptions, thereby enabling accurate and voluminous data generation for large model training. With this system, we create SpeechCraft, a fine-grained bilingual expressive speech dataset. It is distinguished by highly descriptive natural language style prompts, containing approximately 2,000 hours of audio data and encompassing over two million speech clips. Extensive experiments demonstrate that the proposed dataset significantly boosts speech-language task performance in stylist speech synthesis and speech style understanding. © 2024 Owner/Author.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A TSE based design for MMSE and QRD of MIMO systems based on ASIP

引用

High Technology Letters 2023年第2期29卷 166-173页

作者：冯雪林 SHI Jinglin CHEN Yang FU Yanlu ZHANG Qineng XIAO Feng Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing TechnologyChinese Academy of SciencesBeijing 100080P.R.China School of Computer Science and Technology University of Chinese Academy of SciencesBeijing 100049P.R.China Beijing Sylincom Technology Co. Ltd.Beijing 100190P.R.China

A Taylor series expansion(TSE) based design for minimum mean-square error(MMSE) and QR decomposition(QRD) of multi-input and multi-output(MIMO) systems is proposed based on application specific instruction set processor(ASIP), which uses TSE algorithm instead of resource-consuming reciprocal and reciprocal square root(RSR) *** aim is to give a high performance implementation for MMSE and QRD in one programmable platform ***, instruction set architecture(ISA) and the allocation of data paths in single instruction multiple data-very long instruction word(SIMD-VLIW) architecture are provided, offering more data parallelism and instruction parallelism for different dimension matrices and operation ***, multiple level numerical precision can be achieved with flexible table size and expansion order in TSE *** ASIP has been implemented to a 28 nm CMOS process and frequency reaches 800 *** results show that the proposed design provides perfect numerical precision within the fixed bit-width of the ASIP, higher matrix processing rate better than the requirements of 5G system and more rate-area efficiency comparable with ASIC implementations.

关键词： multi-input and multi-output(MIMO) minimum mean-square error(MMSE) QR decomposition(QRD) Taylor series expansion(TSE) application specific instruction set processor(ASIP) instruction set architecture(ISA) single instruction multiple data(SIMD) very long instruction word(VLIW)

来源：评论

学校读者我要写书评

暂无评论

Auto Encoding Neural Process for Multi-interest Recommendation 39

Auto Encoding Neural Process for Multi-interest Recommendati...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Jiang, Yiheng Xu, Yuanbo Yang, Yongjian Yang, Funing Wang, Pengyang Li, Chaozhuo Lab of Mobile Intelligent Computing College of Computer Science and Technology Jilin University China Department of Computer and Information Science The State Key Laboratory of Internet of Things for Smart City University of Macau China Beijing University of Aeronautics and Astronautics China

ISBN: (纸本)157735897X

Multi-interest recommendation constantly aspires to an oracle individual preference modeling approach, one that satisfies the diverse and dynamic properties. Fueled by deep learning technologies, existing neural network (NN)-based recommender systems employ single-point or multi-point interest representation strategies for preference modeling, and boost the recommendation performance with remarkable margins. However, as the parameterized approximate function nature, NN-based methods remain deficiencies regarding the adaptability towards distinctive preference patterns cross different users, and the calibration over individual current intents. In this paper, we revisit multi-interest recommendation with the lens of stochastic process and Bayesian inference. Specifically, we propose to learn a distribution over functions to depict the individual diverse preferences rather than a unified function to approximate preference. Subsequently, the recommendation is equipped with the uncertainty estimation which conforms to the dynamic shifting intent. Along these lines, we establish the connection between multi-interest recommendation and neural processes by proposing NP-Rec, which realizes the flexible multiple interests modeling and uncertainty estimation, simultaneously. Empirical study on 4 real world datasets demonstrates that our NP-Rec attains superior recommendation performances to several state-of-the-art baselines, where the average improvement achieves up to 13.94%. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

VoxInstruct: Expressive Human Instruction-to-Speech Generation with Unified Multilingual Codec Language Modelling 24

VoxInstruct: Expressive Human Instruction-to-Speech Generati...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zhou, Yixuan Qin, Xiaoyu Jin, Zeyu Zhou, Shuoyi Lei, Shun Zhou, Songtao Wu, Zhiyong Jia, Jia Shenzhen International Graduate School Tsinghua University Shenzhen China Department of Computer Science and Technology Tsinghua University Beijing China BNRist Tsinghua University Key Laboratory of Pervasive Computing Ministry of Education Beijing China

ISBN: (纸本)9798400706868

Recent AIGC systems possess the capability to generate digital multimedia content based on human language instructions, such as text, image and video. However, when it comes to speech, existing methods related to human instruction-to-speech generation exhibit two limitations. Firstly, they require the division of inputs into content prompt (transcript) and description prompt (style and speaker), instead of directly supporting human instruction. This division is less natural in form and does not align with other AIGC models. Secondly, the practice of utilizing an independent description prompt to model speech style, without considering the transcript content, restricts the ability to control speech at a fine-grained level. To address these limitations, we propose VoxInstruct, a novel unified multilingual codec language modeling framework that extends traditional text-to-speech tasks into a general human instruction-to-speech task. Our approach enhances the expressiveness of human instruction-guided speech generation and aligns the speech generation paradigm with other modalities. To enable the model to automatically extract the content of synthesized speech from raw text instructions, we introduce speech semantic tokens as an intermediate representation for instruction-to-content guidance. We also incorporate multiple Classifier-Free Guidance (CFG) strategies into our codec language model, which strengthens the generated speech following human instructions. Furthermore, our model architecture and training strategies allow for the simultaneous support of combining speech prompt and descriptive human instruction for expressive speech synthesis, which is a first-of-its-kind attempt. Codes, models and demos are at: https://***/thuhcsi/VoxInstruct. © 2024 Owner/Author.

关键词： C (programming language)

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning-Assisted DementiaNet: A Four Layer Deep CNN for Accurate Alzheimer’s Disease Detection from MRI Images 16th

Transfer Learning-Assisted DementiaNet: A Four Layer Deep CN...

引用

16th International Conference on Brain Informatics, BI 2023

作者： Umme Habiba, Sultana Debnath, Tanoy Islam, Md. Khairul Nahar, Lutfun Shahadat Hossain, Mohammad Basnin, Nanziba Andersson, Karl Khulna University of Engineering & Technology Khulna Bangladesh Department of Computer Science and Engineering Green University of Bangladesh Dhaka Bangladesh Department of Computer Science and Engineering International Islamic University Chittagong Chittagong Bangladesh Department of Computer Science and Engineering University of Chittagong Chittagong Bangladesh Leeds Beckett University Leeds United Kingdom Pervasive and Mobile Computing Laboratory Luleå University of Technology Luleå Sweden

ISBN: (纸本)9783031430749

Alzheimer’s disease is the most common type of dementia and the sixth highest cause of mortality among people over 65. Also, according to statistics, the number of deaths from Alzheimer’s disease has increased dramatically. As a result, early detection of Alzheimer’s disease can improve patient survival chances. Machine learning methods using magnetic resonance imaging have been utilized in the diagnosis of Alzheimer’s disease to help clinicians and speed up the procedure. This research mainly focuses on Alzheimer’s disease detection to overcome previous limitations. We use a publicly available dataset which contains 6400 MRI images. In order to train our dataset, we employ our suggested model, "DementiaNet", using "EfficientNet" as a feature extractor and a Deep CNN as a classifier. In order to capture all the features, this framework uses a small number of convolutional layers, which improves the effectiveness of feature learning and results in a more accurate and reliable output. In addition, to address the issue of data imbalance, we apply data augmentation to enhance the size of the minority class by considering four stages of dementia. Also, in this study, we take advantage of the transfer learning approach with the attachment of "EfficientNet", which allows our model to easily solve the overfitting problem and also extract all the features in a effective way. Finally, we compare our proposed models with those reported in the literature. Our experimental results indicate that our "DementiaNet" model achieve an overall classification accuracy of 97% to detect Alzheimer from brain MRI images, exceeding all other state-of-the-art models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Subjective QoE Assessment for Virtual Reality Cloud-based First-Person Shooter Game

Subjective QoE Assessment for Virtual Reality Cloud-based Fi...

引用

IEEE International Conference on Communications (ICC)

作者： Henrique Souza Rossi Karan Mitra Samuel Larsson Christer Åhlund Irina Cotanis Department of Computer Science Electrical and Space Engineering Pervasive and Mobile Computing Luleå University of Technology Infovista AB Sweden

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

Quality of experience (QoE) is an essential metric for stakeholders to understand how customers perceive the quality of their products or services. Gaming-as-a-Service (GaaS) is a challenging model to deliver efficiently to customers worldwide since it involves the joint force of cloud service providers, network operators, and game developers. The recent move of the cloud gaming (CG) industry to virtual reality (VR) platforms brings the benefits of the cloud to the most immersive quality of service (QoS) and QoE-sensitive VR content. Virtual reality cloud-based gaming (VRCG) necessitates understanding of stochastic broad-band network connections on users' QoE so that stakeholders can deliver quality content by optimizing their services to underlying QoS conditions. Very few studies exist in the literature that study the impact of network QoS on users' QoE for VRCG. This paper presents subjective tests (N=30) and investigates the effect of network-emulated QoS metrics (N=28) on the commercial Nvidia CloudXR service and their impact on the users' perceived QoE while playing Serious Sam VR shooter game. Our findings reveal that QoE was most affected by round trip time (RTT) $\geq 75$ ms or packet loss (PL) $> 6\%$ . Random jitter (RJ) caused QoE degradation for values more significant than one standard deviation, while the combined RTT and PL degraded QoE the most for RTT $\geq 25$ ms and PL $\geq 4\%$ , Finally, based on actual network traffic data between Sweden and various data centers in Europe, we suggest VRCG can be hosted anywhere in these data centers with minimal impact on QoE for wired connections. However, for 4G and 5G networks, high jitter values could pose a challenge to VRCG services.

关键词： Data centers Solid modeling Stochastic processes Quality of service Virtual reality Telecommunication traffic Jitter

来源：评论

学校读者我要写书评

暂无评论

Objective QoE Models for Cloud-Based First Person Shooter Game over mobile Networks

Objective QoE Models for Cloud-Based First Person Shooter Ga...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Henrique Souza Rossi Karan Mitra Christer Åhlund Irina Cotanis Niclas Örgen Per Johansson Department of Computer Science Electrical and Space Engineering Pervasive and Mobile Computing Luleå University of Technology Infovista AB Sweden

mobile cloud gaming (MCG) lets users play cloud games (CG) on mobile devices anywhere via mobile networks. However, the stochastic nature of network quality of service (QoS) can result in varying user quality of experience (QoE). Understanding, modeling, and predicting the impact of mobile networks' QoS on users' QoE is crucial. This helps stakeholders optimize networks, and game developers efficiently create cloud-hosted games provisioned over mobile networks. This paper investigates the impact of QoS on users' QoE and proposes, develops and validates novel models for predicting QoE for MCG in mobile networks using realistic subjective tests. In particular, we propose and develop three QoE models using multiple, polynomial, and non-linear regression. Our results validate that multiple regression (with R 2 =0.79, RMSE=0.45) can model complex relationships between QoS factors that impact QoE. Multiple polynomial regression achieved the overall fit with (R 2 =0.94, RMSE=0.24). Lastly, the non-linear model achieved a good RMSE of 0.24. To select the best model out of the three, we applied the F-test and determined that polynomial regression had the best statistical fit.

关键词：

来源：评论

学校读者我要写书评

暂无评论

QoE Models for Virtual Reality Cloud-based First Person Shooter Game over mobile Networks

QoE Models for Virtual Reality Cloud-based First Person Shoo...

引用

International Conference on Network and Service Management

作者： Henrique Souza Rossi Karan Mitra Christer Åhlund Irina Cotanis Department of Computer Science Electrical and Space Engineering Pervasive and Mobile Computing Luleå University of Technology Infovista AB Sweden

ISBN: (数字)9783903176669

ISBN: (纸本)9798331505158

Virtual reality cloud-based gaming (VRCG) services are becoming widely available on virtual reality (VR) devices delivered over computer networks. VRCG brings users worldwide an extensive catalog of games to play anywhere and anytime. Delivering these gaming services in existing broadband mobile networks is challenging due to their stochastic nature and the user’s perceived Quality of Experience (QoE)’ sensitivity towards them. More research is needed regarding developing effective methods to measure the impact of network QoS factors on users’ QoE in the VRCG context. Therefore, this paper proposes, develops, and validates three novel regression models trained on a real dataset collected via subjective tests (N=30); the dataset contains subjective users’ QoE ratings regarding VR shooter games affected by network conditions (N=28), such as round-trip time (RTT), random jitter (RJ), and packet loss (PL). Our findings reveal that due to the nonlinear relationship of (RTT and RJ) tested together, nonlinear (mean absolute error (MAE)=0.14) and polynomial (MAE=0.15) regression models have the best performance; yet, simple linear regression model (MAE=0.19) is also suitable to predict QoE for VRCG. Further, we found that feature importance depends on the model’s choice (either RTT or RJ). Finally, our models’ prediction of QoE for real-world traffic measurements suggests that mobile network traffic (4G, 5G non-standalone, 5G standalone) provides a 2.5 ≤MOS QoE ≤ 3.0 experience for VRCG, while 4.2 ≤MOS QoE ≤ 4.4 for wired connections, suggesting the need for improvements in the current commercial 5G network deployments to deliver VRCG.

关键词： Solid modeling Cloud computing Sensitivity 5G mobile communication Stochastic processes Virtual reality Games Telecommunication traffic Predictive models Quality of experience

来源：评论

学校读者我要写书评

暂无评论

A Demonstration of ALTRUIST for Conducting QoE Subjective Tests in Immersive Systems

A Demonstration of ALTRUIST for Conducting QoE Subjective Te...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Henrique Souza Rossi Karan Mitra Christer Åhlund Irina Cotanis Department of Computer Science Pervasive and Mobile Computing Electrical and Space Engineering Luleå University of Technology. Infovista AB Sweden

Subjective Quality of Experience (QoE) studies often require setting up complex lab environments to study users' perceptions of the application or service under controlled test conditions. These lab environments must control applications and devices to generate the required test conditions accurately, reliably, repeatedly, and error-free under study. Further, the data collection should be performed on many devices, such as clients and servers, often in real-time, and correctly labelled according to each test condition. To circumvent the complex task of configuring the lab environment and the laborious and error-prone work of data collection, we demonstrate ALTRUIST, a multi-platform tool to conduct subjective tests efficiently. In particular, we present the use of ALTRUIST in two lab setups involving immersive applications such as mobile cloud gaming and virtual reality gaming.

关键词：

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：