检索结果-内蒙古大学图书馆

MGRD: Multigranularity Reconstruction Deviation Modeling for Time Series Anomaly Detection

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Peng, Furong Ma, Rongxin Lu, Xuan Qian, Yuhua Xu, Yong Hu, Zhiguo Wu, Hongtao Institute of Big Data Science and Industry Shanxi University Shanxi Taiyuan030006 China Shanxi University College of Physics and Electronic Engineering Shanxi Taiyuan030006 China Shanxi Intelligent Transportation Institute Company Ltd. Shanxi Taiyuan030032 China Harbin Institute of Technology at Shenzhen School of Computer Science and Technology Peng Cheng Laboratory Shenzhen518055 China

In complex industrial environments, unsupervised time series anomaly detection (AD) with instrumentation and measurement data has significant applications in enhancing reliability and safety. Differentiating noise and anomalies has been a critical topic in AD. Indiscriminately fitting data with deep learning models is susceptible to overfitting. We assume that noise is ubiquitous, while anomaly is manifesting within small segments. To this end, we propose the hypothesis of multigranularity structural deviation in anomalous data from the perspective of suddenness in anomaly signals. Based on this hypothesis, we design a multigranularity time series AD model that builds global- and local-granularity reconstructions for the data and measures the anomalies through the deviation of these reconstructions. The global-granularity reconstruction is computed by all observed data that are equally used for reconstructions, while the local-granularity reconstruction is mainly computed on nearby points. Based on transformers, we propose a multigranularity attention (MGA) module with global and local attention to generate multigranularity outputs. The module employs the Gaussian kernel to control the weights of the nearby points in local attention and applies Kullback-Leibler (KL) divergence to regulate the attention matrices of the two reconstructions from degeneration. The deviation of the two reconstructions can not only detect anomaly data with suddenness changes but also provide intuitive interpretability. Results on five public datasets showcase that our model can obtain 11.38% improvement on average and obtain 38% improvement on the server machine dataset (SMD). © 1963-2012 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

On the Effectiveness of Acoustic BPE in Decoder-Only TTS

arXiv

引用

arXiv 2024年

作者： Li, Bohan Shen, Feiyu Guo, Yiwei Wang, Shuai Chen, Xie Yu, Kai MoE Key Lab of Artificial Intelligence AI Institute X-LANCE Lab Department of Computer Science and Engineering Shanghai Jiao Tong University China Shenzhen Research Institute of Big Data CUHK-Shenzhen China

Discretizing speech into tokens and generating them by a decoder-only model have been a promising direction for text-to-speech (TTS) and spoken language modeling (SLM). To shorten the sequence length of speech tokens, acoustic byte-pair encoding (BPE) has emerged in SLM that treats speech tokens from self-supervised semantic representations as characters to further compress the token sequence. But the gain in TTS has not been fully investigated, and the proper choice of acoustic BPE remains unclear. In this work, we conduct a comprehensive study on various settings of acoustic BPE to explore its effectiveness in decoder-only TTS models with semantic speech tokens. Experiments on LibriTTS verify that acoustic BPE uniformly increases the intelligibility and diversity of synthesized speech, while showing different features across BPE settings. Hence, acoustic BPE is a favorable tool for decoder-only TTS. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Self-Transriber: Few-Shot Lyrics Transcription With Self-Training

Self-Transriber: Few-Shot Lyrics Transcription With Self-Tra...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xiaoxue Gao Xianghu Yue Haizhou Li Department of Electrical and Computer Engineering National University of Singapore Singapore Shenzhen Research Institute of Big Data Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China

The current lyrics transcription approaches heavily rely on supervised learning with labeled data, but such data are scarce and manual labeling of singing is expensive. How to benefit from unlabeled data and alleviate limited data problem have not been explored for lyrics transcription. We propose the first semi-supervised lyrics transcription paradigm, Self-Transcriber, by leveraging on unlabeled data using selftraining with noisy student augmentation. We attempt to demonstrate the possibility of lyrics transcription with a few amount of labeled data. Self-Transcriber generates pseudo labels of the unlabeled singing using teacher model, and augments pseudo-labels to the labeled data for student model update with both self-training and supervised training losses. This work closes the gap between supervised and semi- supervised learning as well as opens doors for few-shot learning of lyrics transcription. Our experiments show that our approach using only 12.7 hours of labeled data achieves competitive performance compared with the supervised approaches trained on 149.1 hours of labeled data for lyrics transcription.

关键词： Training Supervised learning Manuals Signal processing Semisupervised learning data models Acoustics

来源：评论

学校读者我要写书评

暂无评论

Enhancing Large Vision Model in Street Scene Semantic Understanding through Leveraging Posterior Optimization Trajectory

arXiv

引用

arXiv 2025年

作者： Kou, Wei-Bin Lin, Qingfeng Tang, Ming Wang, Shuai Ye, Rongguang Zhu, Guangxu Wu, Yik-Chung Department of Electrical and Electronic Engineering The University of Hong Kong 999077 Hong Kong Shenzhen Research Institute of Big Data Shenzhen China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

To improve the generalization of the autonomous driving (AD) perception model, vehicles need to update the model over time based on the continuously collected data. As time progresses, the amount of data fitted by the AD model expands, which helps to improve the AD model generalization substantially. However, such ever-expanding data is a double-edged sword for the AD model. Specifically, as the fitted data volume grows to exceed the the AD model’s fitting capacities, the AD model is prone to under-fitting. To address this issue, we propose to use a pretrained Large Vision Models (LVMs) as backbone coupled with downstream perception head to understand AD semantic information. This design can not only surmount the aforementioned under-fitting problem due to LVMs’ powerful fitting capabilities, but also enhance the perception generalization thanks to LVMs’ vast and diverse training data. On the other hand, to mitigate vehicles’ computational burden of training the perception head while running LVM backbone, we introduce a Posterior Optimization Trajectory (POT)-Guided optimization scheme (POTGui) to accelerate the convergence. Concretely, we propose a POT Generator (POTGen) to generate posterior (future) optimization direction in advance to guide the current optimization iteration, through which the model can generally converge within 10 epochs. Extensive experiments demonstrate that the proposed method improves the performance by over 66.48% and converges faster over 6 times, compared to the existing state-of-the-art approach. © 2025, CC BY-NC-ND.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hybrid data-Driven SSM for Interpretable and Label-Free mmWave Channel Prediction

引用

IEEE Transactions on Mobile Computing 2025年

作者： Sun, Yiyong He, Jiajun Lin, Zhidi Pu, Wenqiang Yin, Feng So, Hing Cheung The Chinese University of Hon g Kong School of Science and Engineering Shenzhen518172 China BelfastBT3 9DT United Kingdom National University of Singapore Department of Statistics and Data Science 117546 Singapore Shenzhen Research Institute of Big Data Shenzhen518172 China City University of Hong Kong Department of Electrical Engineering Hong Kong

Accurate prediction of mmWave time-varying channels is essential for mitigating the issue of channel aging in highly dynamic scenarios. Existing channel prediction methods have limitations: classical model-based methods often struggle to track highly nonlinear channel dynamics due to limited expert knowledge, while emerging data-driven methods typically require substantial labeled data for effective training and often lack interpretability. To address these issues, this paper proposes a novel hybrid method that integrates a data-driven neural network into a conventional model-based workflow based on a state-space model (SSM), implicitly tracking complex channel dynamics from data without requiring precise expert knowledge. Additionally, a novel unsupervised learning strategy is developed to train the embedded neural network solely with unlabeled data. Theoretical analyses and ablation studies are conducted to interpret the enhanced benefits gained from the hybrid integration. Numerical simulations based on the 3GPP mmWave channel model corroborate the superior prediction accuracy of the proposed method, compared to state-of-the-art methods that are either purely model-based or data-driven. Furthermore, extensive experiments validate its robustness against various challenging factors, including among others severe channel variations. © 2002-2012 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning

引用

IEEE/CAA Journal of Automatica Sinica 2021年第2期8卷 402-411页

作者： Xin Luo Wen Qin Ani Dong Khaled Sedraoui MengChu Zhou the School of Computer Science and Technology Dongguan University of TechnologyDongguan 523808 Hengrui(Chongqing)Artificial Intelligence Research Center Department of Big Data Analyses TechniquesCloudwalkChongqing 401331China the School of Computer Science and Technology Chongqing University of Posts and TelecommunicationsChongqing 400065 Chongqing Engineering Research Center of Big Data Application for Smart Cities Chongqing Institute of Green and Intelligent TechnologyChinese Academy of SciencesChongqing 400714China the Department of Computer and Information Science City College of Dongguan University of TechnologyDongguan 523419China the Department of Electrical and Computer Engineering Faculty of Engineeringand Center of Research Excellence in Renewable Energy and Power SystemsKing Abdulaziz UniversityJeddah 21481Saudi Arabia the Department of Electrical and Computer Engineering New Jersey Institute of TechnologyNewarkNJ 07102 USA the Center of Research Excellence in Renewable Energy and Power Systems King Abdulaziz UniversityJeddah 21481Saudi Arabia

A recommender system(RS)relying on latent factor analysis usually adopts stochastic gradient descent(SGD)as its learning ***,owing to its serial mechanism,an SGD algorithm suffers from low efficiency and scalability when handling large-scale industrial *** at addressing this issue,this study proposes a momentum-incorporated parallel stochastic gradient descent(MPSGD)algorithm,whose main idea is two-fold:a)implementing parallelization via a novel datasplitting strategy,and b)accelerating convergence rate by integrating momentum effects into its training *** it,an MPSGD-based latent factor(MLF)model is achieved,which is capable of performing efficient and high-quality *** results on four high-dimensional and sparse matrices generated by industrial RS indicate that owing to an MPSGD algorithm,an MLF model outperforms the existing state-of-the-art ones in both computational efficiency and scalability.

关键词： big data industrial application industrial data latent factor analysis machine learning parallel algorithm recommender system(RS) stochastic gradient descent(SGD)

来源：评论

学校读者我要写书评

暂无评论

Public Acceptance of Autonomous Vehicle Platooning: A Stated-Preference Survey 24

Public Acceptance of Autonomous Vehicle Platooning: A Stated...

引用

24th COTA International Conference of Transportation Professionals: Resilient, intelligent, Connected, and Lowcarbon Multimodal Transportation, CICTP 2024

作者： Li, Jiankai Ding, Naikan Li, Hongtao Chen, Jiamin Du, Minglin Hu, Ying Aberdeen Institute of Data Science and Artificial Intelligence South China Normal Univ. Foshan China Intelligent Transportation Systems Research Center Wuhan Univ. of Technology Wuhan China Engineering Research Center of Transportation Information and Safety Ministry of Education Wuhan China School of Civil Engineering and Transportation Northeast Forestry Univ. China Harbin Jin Hui End-of-Life Vehicles Recycling Co. Ltd. China

ISBN: (数字)9780784485484

ISBN: (纸本)9780784485484

Autonomous driving platoon control technology has gradually gained widespread attention, but public receptivity to it is shaped by a variety of factors, encompassing cognitive and personal characteristics. This study aims to understand these factors to promote the broader application of autonomous driving platoon control technology. Through questionnaires, we gathered data pertaining to the cognitive characteristics, individual preferences, and personal characteristics of the public. We used a structural equation model to analyze these data to determine the relationships and impact strengths between these factors. Results indicate that cognitive characteristics have a significant positive impact on individual preferences, particularly cognitive characteristics related to safety confidence and technological understanding. Personal characteristics also impact individual preferences. These findings may help policymakers formulate targeted policies by better understanding public needs and concerns. Moreover, it carries substantial implications for the advancement and utilization of autonomous driving platoon control technology, particularly with regard to marketing strategies. © ASCE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Token2vec: A Joint Self-Supervised Pre-Training Framework Using Unpaired Speech and Text

Token2vec: A Joint Self-Supervised Pre-Training Framework Us...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xianghu Yue Junyi Ao Xiaoxue Gao Haizhou Li Department of Electrical and Computer Engineering National University of Singapore Singapore Shenzhen Research Institute of Big Data Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China

Self-supervised pre-training has been successful in both text and speech processing. Speech and text offer different but complementary information. The question is whether we are able to perform a speech-text joint pre-training on unpaired speech and text. In this paper, we take the idea of self-supervised pre-training one step further and propose token2vec, a novel joint pre-training framework for unpaired speech and text based on discrete representations of speech. Specifically, we introduce two modality-specific tokenizers for speech and text. Based on these tokenizers, we convert speech/text sequences into discrete speech/text token sequences consisting of similar language units, thus mitigating the domain mismatch problem and length mismatch problem, which are caused by the distinct characteristics between speech and text. Finally, we feed the discrete speech and text tokens into a modality-agnostic Transformer encoder and pre-train with token-level masking language modeling (tMLM). Experiments show that token2vec is significantly superior to various speech-only pre-training baselines, with up to 17.7% relative WER reduction. Token2vec model is also validated on a non-ASR task, i.e., spoken intent classification, and shows good transferability.

关键词： Analytical models Signal processing Benchmark testing Transformers Acoustics Feeds Task analysis

来源：评论

学校读者我要写书评

暂无评论

Design of approach airspeed monitoring program based on minimum ground speed function 4

Design of approach airspeed monitoring program based on mini...

引用

4th IEEE International Conference on Civil Aviation Safety and Information Technology, ICCASIT 2022

作者： Zhang, Chen Miao, Jiahe Zhu, Liang Zhang, Yuxin Han, Jingru Aviation Safety Analysis and Prevention of Beijing China Academy of Civil Aviation Science and Technology Engineering and Technical Research Center of Civil Beijing China China Academy of Civil Aviation Science & Technology Big Data Analysis and Application Center Beijing China The Second Research Institute CAAC Chengdu China

ISBN: (数字)9781665467667

ISBN: (纸本)9781665467667

Unstable approach is a key issue in the safe flight operation of civil aviation industry. In recent years, the incidence of unstable approach events has increased. An important index to judge the unstable approach is the approach airspeed of the aircraft. CAAC and airlines actively develop monitoring procedures to monitor aircraft approaching airspeed overrun events. However, the accuracy of event monitoring is not high, and there are many false warnings. This paper studies the causes of false warnings, and designs an approach airspeed monitoring program based on the minimum ground speed function. This monitoring program can effectively reduce the occurrence of false warnings and significantly improve the monitoring accuracy of approach airspeed overrun events. Provide more reliable data support for analyzing aircraft unstable approach. © 2022 IEEE.

关键词： Aircraft

来源：评论

学校读者我要写书评

暂无评论

Panicle-Cloud:An Open and Al-Powered Cloud Computing Platform for Quantifying Rice Panicles from Drone-Collected Imagery to Enable the Classification of Yield Production in Rice

引用

Plant Phenomics 2023年第4期5卷 905-916页

作者： Zixuan Teng Jiawei Chen Jian Wang Shuixiu Wu Riqing Chen Yaohai Lin Liyan Shen Robert Jackson Ji Zhou Changcai Yang Digital Fujian Research Institute of Big Data for Agriculture and Forestry College of Computer and Information SciencesFujian Agriculture and Forestry UniversityFuzhou 350002China State Key Laboratory of Crop Genetics&Germplasm Enhancement academy for Advanced Interdisciplinary StudiesNanjing Agricultural UniversityNanjing 210095China Ningxia Academy of Agriculture and Forestry Sciences Yinchuan 750002China College of Mechanical and Electrical Engineering Fujian Agriculture and Forestry UniversityFuzhou 350002China Cambridge Crop Research National Institute of Agricultural Botany(NIAB)Cambridge CB3 OLEUK Key Laboratory of Smart Agriculture and Forestry(Fujian Agriculture and Forestry University) Fujian Province UniversityFuzhou 350002China Center for Agroforestry Mega Data Science School of Future TechnologyFujian Agriculture and Forestry UniversityFuzhou 350002China

Rice(Oryza sativa)is an essential stable food for many rice consumption nations in the world and,thus,the importance to improve its yield production under global climate *** evaluate different rice varieties'yield performance,key yield-related traits such as panicle number per unit area(PNpM^(2))are key indicators,which have attracted much attention by many plant research ***,it is still challenging to conduct large-scale screening of rice panicles to quantify the PNpM^(2)trait due to complex field conditions,a large variation of rice cultivars,and their panicle morphological ***,we present Panicle-Cloud,an open and artificial intelligence(AI)-powered cloud computing platform that is capable of quantifying rice panicles from drone-collected *** facilitate the development of Al-powered detection models,we first established an open diverse rice panicle detection dataset that was annotated by a group of rice specialists;then,we integrated several state-of-the-art deep learning models(including a preferred model called Panicle-AI)into the Panicle-Cloud platform,so that nonexpert users could select a pretrained model to detect rice panicles from their own aerial *** trialed the Al models with images collected at different attitudes and growth stages,through which the right timing and preferred image resolutions for phenotyping rice panicles in the field were ***,we applied the platform in a 2-season rice breeding trial to valid its biological relevance and classified yield production using the platform-derived PNpM^(2)trait from hundreds of rice *** correlation analysis between computational analysis and manual scoring,we found that the platform could quantify the PNpM^(2)trait reliably,based on which yield production was classified with high ***,we trust that our work demonstrates a valuable advance in phenotyping the PNpM^(2)trait in rice,which provides a useful toolkit to enable rice br

关键词： breeding cultivar hundreds

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：