检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Xiao, Haowen Liu, Guanghui Gao, Xinyi Li, Yang Lv, Fengmao Chu, Jielei School of Information Science and Technology Southwest Jiaotong University China School of Computing and Artificial Intelligence Southwest Jiaotong University China College Of Computer Science And Technology Harbin Engineering University China

Self-supervised learning (SSL) has shown remarkable data representation capabilities across a wide range of datasets. However, when applied to real-world datasets with long-tailed distributions, performance on multiple downstream tasks degrades significantly. Recently, the community has begun to focus more on self-supervised long-tailed learning. Some works attempt to transfer temperature mechanisms to self-supervised learning or use category-space uniformity constraints to balance the representation of different categories in the embedding space to fight against long-tail distributions. However, most of these approaches focus on the joint optimization of all samples in the dataset or on constraining the category distribution, with little attention given to whether each individual sample is optimally guided during training. To address this issue, we propose Temperature Auxiliary Sample-level Encouragement (TASE). We introduce pseudo-labels into self-supervised long-tailed learning, utilizing pseudo-label information to drive a dynamic temperature and re-weighting strategy. Specifically, We assign an optimal temperature parameter to each sample. Additionally, we analyze the lack of quantity awareness in the temperature parameter and use re-weighting to compensate for this deficiency, thereby achieving optimal training patterns at the sample level. Comprehensive experimental results on six benchmarks across three datasets demonstrate that our method achieves outstanding performance in improving long-tail recognition, while also exhibiting high robustness. © 2024, CC BY.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Keyword Spotting from Mixed Speech

arXiv

引用

arXiv 2024年

作者： Yuan, Junming Shi, Ying Li, LanTian Wang, Dong Hamdulla, Askar School of Computer Science and Technology Xinjiang University China School of Artificial Intelligence Beijing University of Posts and Telecommunications China Center for Speech and Language Technologies BNRist Tsinghua University China School of Computer Science and Technology Harbin Institute of Technology China

Few-shot keyword spotting (KWS) aims to detect unknown keywords with limited training samples. A commonly used approach is the pre-training and fine-tuning framework. While effective in clean conditions, this approach struggles with mixed keyword spotting – simultaneously detecting multiple keywords blended in an utterance, which is crucial in real-world applications. Previous research has proposed a Mix-Training (MT) approach to solve the problem, however, it has never been tested in the few-shot scenario. In this paper, we investigate the possibility of using MT and other relevant methods to solve the two practical challenges together: few-shot and mixed speech. Experiments conducted on the LibriSpeech and Google Speech Command corpora demonstrate that MT is highly effective on this task when employed in either the pre-training phase or the fine-tuning phase. Moreover, combining SSL-based large-scale pre-training (HuBert) and MT fine-tuning yields very strong results in all the test conditions. © 2024, CC BY-NC-SA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hand Function Assessment using computer Vision for Hand Rehabilitation

Hand Function Assessment using Computer Vision for Hand Reha...

引用

artificial intelligence (CAI), IEEE Conference on

作者： Kamal Debanga Raj Neog H Pallab Jyoti Dutta M. K. Bhuyan Mehta Family School of Data Science and Artificial Intelligence Department of Electronics and Electrical Engineering Indian Institute of Technology Guwahati Assam India

ISBN: (数字)9798350354096

ISBN: (纸本)9798350354102

Individuals who suffer a stroke often experience impaired hand function. The normal function can be restored by hand rehabilitation exercises with constant monitoring and assessment of motor recovery progress. However, this poses a challenge to patients due to regular hospital visits and physiotherapy sessions. We address this inconvenience by developing a reliable vision-based hand rehabilitation system that guides patients to perform rehabilitation exercises, such as the Sollerman Hand Function Test and Box and Block Test. The patients are asked to perform exercises based on instructions shown on a computer screen and the system tracks finger and hand to estimate performance scores. This scoring provides feedback to the patients to monitor their progress and improve their hand function gradually. Experiments conducted with ten males and five females highlight that the proposed vision-based hand rehabilitation exercises are effective in improving hand functionalities.

关键词： Solid modeling Three-dimensional displays Hospitals Computational modeling Lighting Writing Stroke (medical condition)

来源：评论

学校读者我要写书评

暂无评论

artificial intelligence Cultivation: Transforming Agriculture for a Smart and Sustainable Future

Artificial Intelligence Cultivation: Transforming Agricultur...

引用

Intelligent data Communication Technologies and Internet of Things (IDCIoT), International Conference on

作者： Bhushan Fulkar Samir Mendhe Pawan Patil Department of Artificial Intelligence and Data science Faculty of Engineering and Technology DMIHER(DU) Wardha Maharashtra India Computer Science and Design Faculty of Engineering and Technology DMIHER(DU) Wardha Maharashtra India Artificial Intelligence and Data science Faculty of Engineering and Technology DMIHER(DU) Wardha Maharashtra India

The growing global population is posing unprecedented challenges for the agriculture industry, requiring it to produce more food while also addressing resource constraints and environmental concerns. artificial intelligence (AI) has changed the game in agriculture by offering creative solutions to increase productivity, optimize resource use, and promote sustainability. artificial intelligence (AI) has been used in the agriculture sector more and more recently. A few of the problems the industry faces in attempting to maximize its produce are improper soil treatment, disease and pest infestation, the use of driverless tractors, big data requirements, and the knowledge gap between farmers and technology. This work provides an overview of the applications of artificial intelligence in crop, weed, and disease management. The proposed work aims to review various AI techniques, including artificial neural networks (ANN) and fuzzy logic (FL).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised PolSAR image classification method based on contrastive learning 6

Semi-supervised PolSAR image classification method based on ...

引用

6th International Conference on artificial intelligence and Pattern Recognition, AIPR 2023

作者： Hua, Wenqiang Sun, Nan Liu, Lin The Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an Key Laboratory of Big Data and Intelligent Computing School of Computer Science and Technology Xi'an University of Posts and Telecommunications Xi'an China School of Computer Technology Xi'an University of Posts and Telecommunications Xi'an China

ISBN: (纸本)9798400707674

Polarimetric synthetic aperture radar (PolSAR) image classification has important application value and a wide range of application scenarios in many fields. Supervised classification methods, which need to use a large amount of labeled data to improve classification accuracy, have achieved satisfactory classification performance. However, it is difficult and expensive to obtain sufficient labeled data for PolSAR images. The problem to be solved is how to obtain high classification accuracy with a limited number of labeled samples. Semi-supervised PolSAR image classification method based on contrastive learning is proposed. Firstly, a supervised PolSAR image classification method based on meta-learning is introduced. Prototypical network based on meta-learning ideas is used as the basic classification model, and a prototypical representation is created for each category. Classification is performed by computing the similarity between the prototypical representation and samples to be classified. Secondly, a method of training feature extractor based on contrastive learning is designed. By maintaining the consistency of features between the same data and distinguishing the differences between different data after data augmentation, the method enables the feature extractor to obtain accurate feature representation. By co-training with meta-learning, so as to improve the classification performance. Finally, experiments on real PolSAR data prove the effectiveness of the proposed method. © 2023 ACM.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Learning Semantical Dynamics and SpatioTemporal Collaboration for Human Pose Estimation in Video

arXiv

引用

arXiv 2025年

作者： Feng, Runyang Chen, Haoming School of Artificial Intelligence Jilin University Jilin Changchun130015 China School of Computer Science and Technology East China Normal University Shanghai200062 China

Temporal modeling and spatio-temporal collaboration are pivotal techniques for video-based human pose estimation. Most state-of-the-art methods adopt optical flow or temporal difference, learning local visual content correspondence across frames at the pixel level, to capture motion dynamics. However, such a paradigm essentially relies on localized pixel-to-pixel similarity, which neglects the semantical correlations among frames and is vulnerable to image quality degradations (e.g. occlusions or blur). Moreover, existing approaches often combine motion and spatial (appearance) features via simple concatenation or summation, leading to practical challenges in fully leveraging these distinct modalities. In this paper, we present a novel framework that learns multi-level semantical dynamics and dense spatio-temporal collaboration for multi-frame human pose estimation. Specifically, we first design a Multi-Level Semantic Motion Encoder using a multi-masked context and pose reconstruction strategy. This strategy stimulates the model to explore multi-granularity spatiotemporal semantic relationships among frames by progressively masking the features of (patch) cubes and frames. We further introduce a Spatial-Motion Mutual Learning module which densely propagates and consolidates context information from spatial and motion features to enhance the capability of the model. Extensive experiments demonstrate that our approach sets new state-of-the-art results on three benchmark datasets, PoseTrack2017, PoseTrack2018, and PoseTrack21. © 2025, CC BY-NC-ND.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

A Multimodal Personality Detection Method With Behavioral Action Feature and Association Features *

A Multimodal Personality Detection Method With Behavioral Ac...

引用

Smart World Congress (SWC), IEEE

作者： Hao Yang Ruicong Zhi Rongquan Wang Xianyu Xu Lin Wei School of Computer and Artificial Intelligence Zhengzhou University Zhengzhou China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China College of Artificial Intelligence Tianjin University of Science and Technology Tianjin China Flight Technology College Civil Aviation Flight University of China Guanghan China

ISBN: (数字)9798331520861

ISBN: (纸本)9798331520878

Multimodal personality detecting aims to identify human personality traits from videos in computer vision. Existing methods mine different modal features separately and fuse them for the Big Five personality detection task, neglecting the mining of multimodal correlation features and behavioral action characteristics. Therefore, we propose a novel multimodal feature learning method for analyzing the personality traits of individuals in short videos. Firstly, we design an Image-Audio Association Feature extraction method (IAAF) to capture the correlation features between images and audio. This method utilizes existing pre-trained models to separately mine image and audio features and calculates the dot product between the two modal features as the association feature. Considering the role of behavioral action features in personality prediction tasks, we utilize pre-trained 3D-ResNet34 as a feature extractor to mine this feature. Meanwhile, We use CLIP to extract the image-personality descriptions feature between images and personality descriptors. This paper uses a pre-trained Roberta to extract text features to enhance the method’s robustness. Finally, the proposed method is tested on two benchmark datasets, and it outperforms existing state-of-the-art methods.

关键词： Representation learning computer vision Correlation Predictive models Feature extraction Prediction algorithms Robustness Multilingual Convolutional neural networks Videos

来源：评论

学校读者我要写书评

暂无评论

Heavy-Duty Emission Prediction Model Using Wavelet Features and ResNet 21st

Heavy-Duty Emission Prediction Model Using Wavelet Features ...

引用

21st Asia Simulation Conference, AsiaSim 2022

作者： Wang, Ruibin Xia, Xiushan Xu, Zhenyi School of Computer Science and Technology Anhui University Hefei230039 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China Institute of Advanced Technology University of Science and Technology of China Hefei230088 China

ISBN: (纸本)9789811991974

To solve the problem that the COPERT model cannot be applied to heavy-duty diesel vehicle OBD monitoring data to obtain accurate NOx emission factors, we propose a ResNet amendment model based on historical driving time-frequency wavelet features. First, for multiple consecutive driving segments of the actual collected diesel vehicle single-vehicle trip data, the attributes with high correlation with the actual OBD emission factors were obtained using Spearman rank correlation analysis based on the data volume. Then, the highly correlated attributes and the emission factors predicted by the COPERT model were combined by constructing the historical information matrix and using the continuous wavelet transform for the time-frequency representation;Finally, the time-frequency representation was used as the input to complete the amendment of the COPERT model on the ResNet50 network. The experimental results show that the proposed method can effectively amend the COPERT model so that the COPERT model can be applied to OBD data to achieve accurate prediction of NOx emission factors. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Diesel engines

来源：评论

学校读者我要写书评

暂无评论

Country profiling through IAEA Milestones Approach: A comprehensive analysis of the Moroccan nuclear infrastructure

引用

IFAC-PapersOnLine 2024年第13期58卷 751-756页

作者： H. Housni N. Amrous N. Daoudi M.J. Malzi Laboratory: Mediation Information Knowledge Society. Research Team: Competitive Intelligence Analysis and Strategy School of Information Sciences Rabat Morocco Laboratory LyRIC : Computer Science Data Science and Artificial Intelligence. ITQAN Research Team: Innovative Technologies data Quality and Artificial iNtelligence School of Information Sciences Rabat Morocco Research Laboratory in Energy Environment and Resource Economics (GREER) Cadi Ayyad University Marrakech Morocco

Morocco’s exploration of nuclear energy aligns with climate goals and national energy ambitions, providing a low-carbon alternative to fossil fuels. Despite the adoption of renewable energies, Morocco still needs to enhance its energy resilience. This article explores Morocco’s preparedness in the case of considering the introduction of nuclear energy to support the energy mix and mitigate the issue of energy dependence. This study employs the International Atomic Energy Agency (IAEA) milestones approach to analyze Morocco’s nuclear readiness and address existing literature gaps. By offering this analysis, the article aims to inform appropriate policy decisions regarding the potential introduction of nuclear energy. Results indicate the nation’s progression into Phase 1, advancing toward a nuclear program decision. Achieving milestone 3 and completing Phase 3 necessitate essential competencies and extensive infrastructure development.

关键词： Nuclear Energy Morocco Milestones Approach Country profile

来源：评论

学校读者我要写书评

暂无评论

Unleashing the Power of Continual Learning on Non-Centralized Devices: A Survey

引用

IEEE Communications Surveys and Tutorials 2025年

作者： Li, Yichen Wang, Haozhao Xu, Wenchao Xiao, Tianzhe Liu, Hong Tu, Minzhu Wang, Yuying Yang, Xin Zhang, Rui Yu, Shui Guo, Song Li, Ruixuan Huazhong University of Science and Technology School of Computer Science and Technology Wuhan China The Hong Kong Polytechnic University Department of Computing Hong Kong Beijing China Soochow University School of Computer Science and Technology Suzhou China Southwestern University of Finance and Economics School of Computing and Artificial Intelligence Chengdu China University of Technology Sydney School of Computer Science Sydney Australia The Hong Kong University of Science and Technology Department of Computer Science and Engineering Hong Kong

Non-Centralized Continual Learning (NCCL) has become an emerging paradigm for enabling distributed devices such as vehicles and servers to handle streaming data from a joint non-stationary environment. To achieve high reliability and scalability in deploying this paradigm in distributed systems, it is essential to overcome challenges stemming from both spatial and temporal dimensions, manifesting as distribution shifts, catastrophic forgetting, heterogeneity, and privacy issues. This survey focuses on a comprehensive examination of the development of the non-centralized continual learning algorithms and the real-world deployment across distributed devices. We begin with an introduction to the background and fundamentals of non-centralized learning and continual learning. Then, we review existing solutions from three levels to represent how existing techniques alleviate the catastrophic forgetting and distribution shift. Additionally, we delve into the various types of heterogeneity issues, security, and privacy attributes, as well as real-world applications across three prevalent scenarios. Furthermore, we establish a large-scale benchmark to revisit this problem and analyze the performance of the state-of-the-art NCCL approaches. Finally, we discuss the important challenges and future research directions in NCCL. © 2021 IEEE.

关键词： Learning algorithms

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：