检索结果-南通市图书馆

Robust video question answering via contrastive cross-modality representation learning

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(information sciences) 2024年第10期67卷 211-226页

作者： Xun YANG Jianming ZENG Dan GUO Shanshan WANG Jianfeng DONG Meng WANG School of information science and Technology University of Science and Technology of China institute of Artificial Intelligence Hefei Comprehensive National Science Center School of Computer science and information Engineering Hefei University of Technology institutes of Physical science and information Technology Anhui University School of Computer science and Technology Zhejiang Gongshang University

Video question answering(VideoQA) is a challenging yet important task that requires a joint understanding of low-level video content and high-level textual semantics. Despite the promising progress of existing efforts, recent studies revealed that current VideoQA models mostly tend to over-rely on the superficial correlations rooted in the dataset bias while overlooking the key video content, thus leading to unreliable results. Effectively understanding and modeling the temporal and semantic characteristics of a given video for robust VideoQA is crucial but, to our knowledge, has not been well investigated. To fill the research gap, we propose a robust VideoQA framework that can effectively model the cross-modality fusion and enforce the model to focus on the temporal and global content of videos when making a QA decision instead of exploiting the shortcuts in datasets. Specifically, we design a self-supervised contrastive learning objective to contrast the positive and negative pairs of multimodal input, where the fused representation of the original multimodal input is enforced to be closer to that of the intervened input based on video perturbation. We expect the fused representation to focus more on the global context of videos rather than some static keyframes. Moreover, we introduce an effective temporal order regularization to enforce the inherent sequential structure of videos for video representation. We also design a Kullback-Leibler divergence-based perturbation invariance regularization of the predicted answer distribution to improve the robustness of the model against temporal content perturbation of videos. Our method is model-agnostic and can be easily compatible with various VideoQA backbones. Extensive experimental results and analyses on several public datasets show the advantage of our method over the state-of-the-art methods in terms of both accuracy and robustness.

关键词： video question answering cross-modality fusion contrastive learning cross-media reasoning

Unleashing potentials with deep learning: decoding the complex events for distributed fiber optic sensing applications

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(information sciences) 2024年第5期67卷 337-338页

作者： Yujiao LI Liqin HU Kuanglu YU institute of information science School of Computer and Information Technology Beijing Jiaotong University Beijing Key Laboratory of Advanced information science and Network Technology Beijing Jiaotong University

After decades of research, distributed fiber optic sensors have become essential tools for precise long-distance measurements. They are applied across various sectors such as railway transportation, gas and petroleum, and smart grids. Signal processing, particularly in phase-sensitive optical time domain reflectometry(Φ-OTDR) event recognition and classification, is a primary research focus.

关键词： Deep learning

Self-supervised recalibration network for person re-identification

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第1期31卷 163-178页

作者： Shaoqi Hou Zhiming Wang Zhihua Dong Ye Li Zhiguo Wang Guangqiang Yin Xinzhong Wang Shenzhen institute of information Technology Shenzhen518172China University of Electronic science and Technology of China Chengdu611731China Kash institute of Electronics and information Industry Kashi844099China

The attention mechanism can extract salient features in images,which has been proved to be effective in improving the performance of person re-identification(Re-ID).However,most of the existing attention modules have the following two shortcomings:On the one hand,they mostly use global average pooling to generate context descriptors,without highlighting the guiding role of salient information on descriptor generation,resulting in insufficient ability of the final generated attention mask representation;On the other hand,the design of most attention modules is complicated,which greatly increases the computational cost of the *** solve these problems,this paper proposes an attention module called self-supervised recalibration(SR)block,which introduces both global and local information through adaptive weighted fusion to generate a more refined attention *** particular,a special"Squeeze-Excitation"(SE)unit is designed in the SR block to further process the generated intermediate masks,both for nonlinearizations of the features and for constraint of the resulting computation by controlling the number of ***,we combine the most commonly used Res Net-50 to construct the instantiation model of the SR block,and verify its effectiveness on multiple Re-ID datasets,especially the mean Average Precision(m AP)on the Occluded-Duke dataset exceeds the state-of-the-art(SOTA)algorithm by 4.49%.

关键词： Person re-identification Attention mechanism Global information Local information Adaptive weighted fusion

Categorizing methods for integrating machine learning with executable specifications

在线全文

学校读者我要写书评

暂无评论

science China(information sciences) 2024年第1期67卷 5-19页

作者： David HAREL Raz YERUSHALMI Assaf MARRON Achiya ELYASAF Department of Computer science and Applied Mathematics Weizmann Institute of Science Department of Software and information Systems Engineering Ben-Gurion University of the Negev

Deep learning(DL),which includes deep reinforcement learning(DRL),holds great promise for carrying out real-world tasks that human minds seem to cope with quite *** promise is already delivering extremely impressive results in a variety of ***,while DL-enabled systems achieve excellent performance,they are far from *** has been demonstrated,in several domains,that DL systems can err when they encounter cases they had not hitherto ***,the opacity of the produced agents makes it difficult to explain their behavior and ensure that they adhere to various requirements posed by human *** the other end of the software development spectrum of methods,behavioral programming(BP) facilitates orderly system development using self-standing executable modules aligned with how humans intuitively describe desired system *** this paper,we elaborate on different approaches for combining DRL with BP and,more generally,machine learning(ML) with executable specifications(ES).We begin by defining a framework for studying the various approaches,which can also be used to study new emerging approaches not covered *** then briefly review state-of-the-art approaches to integrating ML with ES,continue with a focus on DRL,and then present the merits of integrating ML with *** conclude with guidelines on how this categorization can be used in decision making in system development,and outline future research challenges.

关键词： machine learning artificial intelligence grey box learning domain knowledge rules behavioral programming deep reinforcement learning survey

Steering the energy sharing of electrons in nonsequential double ionization with orthogonally polarized two-color field

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第8期33卷 248-252页

作者：樊光琦杨志杰孙烽豪郑金梅韩云天黄明谦刘情操 College of science Harbin Institute of TechnologyWeihai 264209China School of information science and Engineering Harbin Institute of TechnologyWeihai 264209China Harbin institute of Technology Weihai 264209China

Using the semiclassical ensemble model,the dependence of relative amplitude for the recollision dynamics in nonsequential double ionization(NSDI)of neon atom driven by the orthogonally polarized two-color field(OTC)laser field is theoretically *** the dynamics in two typical collision pathways,recollision-impact-ionization(RII)and recollisionexcitation with subsequent ionization(RESI),is systematically *** results reveal that the V-shaped structure in the correlated momentum distribution is mainly caused by the RII mechanism when the relative amplitude of the OTC laser field is zero,and the first ionized electrons will quickly skim through the nucleus and share few energy with the second *** the relative amplitude increases,the V-shaped structure gradually disappears and electrons are concentrated on the diagonal in the electron correlation spectrum,indicating that the energy sharing after electrons collision is symmetric for OTC laser fields with large relative *** studies show that changing the relative amplitude of the OTC laser field can efficiently control the electron–electron collisions and energy exchange efficiency in the NSDI process.

关键词： nonsequential double ionization correlated electron–electron momentum distribution energy sharing of electrons orthogonally polarized two-color field laser field semiclassical ensemble models

Thickness dependence of superconductivity in layered GeP_(5)

在线全文

学校读者我要写书评

暂无评论

Rare Metals 2024年第3期43卷 1323-1328页

作者： Lin Wu Ke-Meng Yang Wei Gan Rui-Chun Xiao Ya-Jie Dai Xi Tang Hong Lin Hui Han Hui Li information Materials and Intelligent Sensing Laboratory of Anhui Province Institute of Physical Science and Information TechnologyAnhui UniversityHefei230601China

Understanding the relationship between dimensionality and superconductivity has attracted substantial research *** this work,we report the effective modulation of superconductivity in GeP_(5)through *** superconducting phase transition temperature improves by~10%as the thickness of GeP_(5)decreases from bulk to~96 nm,behaving as a unique negative relationship between the superconducting phase transition temperatureand thickness.

关键词： superconductivity dimensionality

TransTM:A device-free method based on time-streaming multiscale transformer for human activity recognition

在线全文

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第2期32卷 619-628页

作者： Yi Liu Weiqing Huang Shang Jiang Bobai Zhao Shuai Wang Siye Wang Yanfang Zhang institute of information Engineering Chinese Academy of sciences School of Cyber SecurityUniversity of Chinese Academy of SciencesBeijingChina School of information Management Beijing Information Science and Technology UniversityBeijingChina

RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy *** RFID-based HAR methods use modeling,CNN,or LSTM to extract features ***,they have shortcomings:1)requiring complex hand-crafted data cleaning processes and 2)only addressing single-person activity recognition based on specific RF *** solve these problems,this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called *** model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without ***,we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes singlehuman activities and human-to-human *** with existing CNN-and LSTM-based methods,the Transformer-based method has more data fitting power,generalization,and ***,using RF signals,our method achieves an excellent classification effect on human behaviorbased classification *** results on the actual RFID datasets show that this model achieves a high average recognition accuracy(99.1%).The dataset we collected for detecting RFID-based indoor human activities will be published.

关键词： Human activity recognition RFID Transformer

Nonparametric Statistical Feature Scaling Based Quadratic Regressive Convolution Deep Neural Network for Software Fault Prediction

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第3期78卷 3469-3487页

作者： Sureka Sivavelu Venkatesh Palanisamy School of Computer science Engineering and information Systems Vellore Institute of TechnologyVellore632014India

The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two sta

关键词： Software defect prediction feature selection nonparametric statistical Torgerson-Gower scaling technique quadratic censored regressive convolution deep neural network softstep activation function nelder-mead method

2P3FL:A Novel Approach for Privacy Preserving in Financial Sectors Using Flower Federated Learning

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & sciences 2024年第8期140卷 2035-2051页

作者： Sandeep Dasari Rajesh Kaluri School of Computer science Engineering and information Systems Vellore Institute of TechnologyVellore632014India

The increasing data pool in finance sectors forces machine learning(ML)to step into new *** data has significant financial implications and is *** users data from several organizations for various banking services may result in various intrusions and privacy *** a result,this study employs federated learning(FL)using a flower paradigm to preserve each organization’s privacy while collaborating to build a robust shared global ***,diverse data distributions in the collaborative training process might result in inadequate model learning and a lack of *** address this issue,the present paper proposes the imple-mentation of Federated Averaging(FedAvg)and Federated Proximal(FedProx)methods in the flower framework,which take advantage of the data locality while training and guaranteeing global *** improves the privacy of the local *** analysis used the credit card and Canadian institute for Cybersecurity Intrusion Detection Evaluation(CICIDS)***,recall,and accuracy as performance indicators to show the efficacy of the proposed strategy using FedAvg and *** experimental findings suggest that the proposed approach helps to safely use banking data from diverse sources to enhance customer banking services by obtaining accuracy of 99.55%and 83.72%for FedAvg and 99.57%,and 84.63%for FedProx.

关键词： Federated learning FedAvg FedProx flower framework privacy preservation financial sectors