检索结果-南通市图书馆

Span-based joint entity and relation extraction augmented with sequence tagging mechanism

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第5期67卷 84-98页

作者： Bin JI Shasha LI Hao XU Jie YU Jun MA Huijun LIU Jing YANG College of Computer National University of Defense Technology

Span-based joint extraction simultaneously conducts named entity recognition(NER) and relation extraction(RE) in a text span form. However, since previous span-based models rely on span-level classifications, they cannot benefit from token-level label information, which has been proven advantageous for the task. In this paper, we propose a sequence tagging augmented span-based network(STSN), a spanbased joint model that can make use of token-level label information. In STSN, we construct a core neural architecture by deep stacking multiple attention layers, each of which consists of three basic attention *** the one hand, the core architecture enables our model to learn token-level label information via the sequence tagging mechanism and then uses the information in the span-based joint extraction; on the other hand, it establishes a bi-directional information interaction between NER and RE. Experimental results on three benchmark datasets show that STSN consistently outperforms the strongest baselines in terms of F1,creating new state-of-the-art results.

关键词： joint extraction named entity recognition relation extraction span sequence tagging mechanism

High mechanical strength Si anode synthesis with interlayer bonded expanded graphite structure for lithium-ion batteries

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Energy Chemistry 2024年第6期93卷 253-263,I0007页

作者： Wenhui Lai Jong Hak Lee Lu Shi Yuqing Liu Yanhui Pu Yong Kang Ong Carlos Limpo Ting Xiong Yifan Rao Chorng Haur Sow Barbaros Ozyilmaz Department of Material Science&Engineering National University of SingaporeSingapore 117575Singapore Center for Advanced 2D Materials National University of SingaporeSingapore 117546Singapore Department of Physics National University of SingaporeSingapore 117551Singapore Institute forFunctional Intelligent Materials National University of SingaporeSingapore 117544Singapore

Despite advancements in silicon-based anodes for high-capacity lithium-ion batteries,their widespread commercial adoption is still hindered by significant volume expansion during cycling,especially at high active mass loadings crucial for practical *** root of these challenges lies in the mechanical instability of the material,which subsequently leads to the structural failure of the ***,we present a novel synthesis of a composite combining expanded graphite and silicon *** composite features a unique interlayer-bonded graphite structure,achieved through the application of a modified spark plasma sintering ***,this innovative structure not only facilitates efficient ion and electron transport but also provides exceptional mechanical strength(Vickers hardness:up to658 MPa,Young's modulus:11.6 GPa).This strength effectively accommodates silicon expansion,resulting in an impressive areal capacity of 2.9 mA h cm^(-2)(736 mA h g^(-1)) and a steady cycle life(93% after 100cycles).Such outsta nding performance is paired with features appropriate for large-scale industrial production of silicon batteries,such as active mass loading of at least 3.9 mg cm^(-2),a high-tap density electrode material of 1.68 g cm^(-3)(secondary clusters:1.12 g cm^(-3)),and a production yield of up to 1 kg per day.

关键词： Lithium-ion battery Silicon anode Spark plasma sintering Interlayer bonding Mechanical strength Tap density

Adaptive and augmented active anomaly detection on dynamic network traffic streams

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2024年第3期25卷 446-460页

作者： Bin LI Yijie WANG Li CHENG national Key Laboratory of Parallel and Distributed Computing College of ComputerNational University of Defense TechnologyChangsha 410073China College of System Engineering National University of Defense TechnologyChangsha 410073China

Active anomaly detection queries labels of sampled instances and uses them to incrementally update the detection model,and has been widely adopted in detecting network ***,existing methods cannot achieve desirable performance on dynamic network traffic streams because(1)their query strategies cannot sample informative instances to make the detection model adapt to the evolving stream and(2)their model updating relies on limited query instances only and fails to leverage the enormous unlabeled instances on *** address these issues,we propose an active tree based model,adaptive and augmented active prior-knowledge forest(A3PF),for anomaly detection on network trafic streams.A prior-knowledge forest is constructed using prior knowledge of network attacks to find feature subspaces that better distinguish network anomalies from normal *** one hand,to make the model adapt to the evolving stream,a novel adaptive query strategy is designed to sample informative instances from two aspects:the changes in dynamic data distribution and the uncertainty of *** the other hand,based on the similarity of instances in the neighborhood,we devise an augmented update method to generate pseudo labels for the unlabeled neighbors of query instances,which enables usage of the enormous unlabeled instances during model *** experiments on two benchmarks,CIC-IDS2017 and UNSW-NB15,demonstrate that A3PF achieves significant improvements over previous active methods in terms of the area under the receiver operating characteristic curve(AUC-ROC)(20.9%and 21.5%)and the area under the precision-recall curve(AUC-PR)(44.6%and 64.1%).

关键词： Active anomaly detection Network traffic streams Pseudo labels Prior knowledge of network attacks

Multiscale observation in wide-spatial radar surveillance based on coherent FDA design

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第2期67卷 232-252页

作者： Lei YU Feng HE Yi SU College of Electronics Science National University of Defense Technology

Wide-spatial radar surveillance missions are challenging tasks, requiring an increased power budget and agility in transmission aimed at extracting information from multiple targets in different environments. These requirements necessitate high transmitting degrees of freedom(DOF) to achieve the objective of multiscale observation for specific tasks in specialized regions of interest. Herein, we exploit the multiscale observation ability in wide-spatial radar surveillance based on frequency diverse array(FDA)radar. The proposed method facilitates spatial anisotropic control of multiple radar resources, including the transmitting waveforms, beampattern, and bandwidth. By utilizing a coherent FDA radar, we offer principles for the selection of baseband waveforms, in addition to the quantitative design of the beampattern gain and optimal bandwidth from the perspective of detection. The feasibility of the proposed method is validated through numerical experiments, thus indicating the potential in wide-spatial radar surveillance. Moreover,this work can be regarded as a preliminary attempt to gauge the efficacy of the computational array, a novel academic concept.

关键词： multi-scale observation wide-spatial surveillance frequency diverse array (FDA) beampattern synthesis frequency diversity gain computational array

Beamforming design for covert broadcast communication with hidden adversary

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6期67卷 384-398页

作者： Weiyu CHEN Haiyang DING Shilian WANG Junshan LUO Fengkui GONG College of Electronic Science and Technology National University of Defense Technology College of Information and Communication National University of Defense Technology State Key Laboratory of Integrated Service Networks Xidian University

This paper explores covert broadcast communication in a challenging situation in which the transmitter, Alice, faces uncertainty regarding the position of the passive adversary, Willie. In this situation,controlling the signal exposure becomes difficult. Specifically, this paper focuses on the cases where Alice can delimit the suspicious areas where Willie may be located. The suspicious areas can have arbitrary number and shapes. Under the Rician fading model, the analytical expression of Willie's detection performance is derived. Then, the transmit power, the beamforming direction, and the number of channel uses are jointly designed to minimize the optimal Willie's detection performance among all the possible positions of Willie, on the premise of satisfying receivers' quality-of-service requirements. To tackle the formulated knotty problem with an infinite number of irregular and discontinuous Willie's possible positions, the continuous nature of the antenna pattern is exploited to convert the original problem into the one with a finite number of possible positions, by sampling the suspicious areas. Then, different algorithms are developed to address the resultant non-convex optimization problem for single-antenna Willie and multi-antenna Willie, respectively. Numerical results demonstrate significant performance gains of the proposed scheme over the widely-adopted maximum ratio transmission scheme, which indicates that simply adjusting the main lobe towards the receiver is indeed far away from optimum and shows the importance of leveraging Alice's limited knowledge of Willie's position.

关键词： low-probability-of-detection communication multiple antennas covertness performance uncertain position parameters optimization

同方期刊数据库

Transformer in reinforcement learning for decision-making:a survey

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2024年第6期25卷 763-790页

作者： Weilin YUAN Jiaxing CHEN Shaofei CHEN Dawei FENG Zhenzhen HU Peng LI Weiwei ZHAO College of Information and Communication National University of Defense TechnologyWuhan 430014China College of Intelligence Science and Technology National University of Defense TechnologyChangsha 410072China Laboratory for Parallel and Distributed Processing National University of Defense TechnologyChangsha 410072China

Reinforcement learning(RL)has become a dominant decision-making paradigm and has achieved notable success in many real-world ***,deep neural networks play a crucial role in unlocking RL’s potential in large-scale decision-making *** by current major success of Transformer in natural language processing and computer vision,numerous bottlenecks have been overcome by combining Transformer with RL for *** paper presents a multiangle systematic survey of various Transformer-based RL(TransRL)models applied in decision-making tasks,including basic models,advanced algorithms,representative implementation instances,typical applications,and known *** work aims to provide insights into problems that inherently arise with the current RL approaches,and examines how we can address them with better TransRL *** our knowledge,we are the first to present a comprehensive review of the recent Transformer research developments in RL for *** hope that this survey provides a comprehensive review of TransRL models and inspires the RL community in its pursuit of future *** keep track of the rapid TransRL developments in the decision-making domains,we summarize the latest papers and their open-source implementations at https://***/williamyuanv0/Transformer-in-Reinforcement-Learning-for-Decision-Making-A-Survey.

关键词： Transformer Reinforcement learning(RL) Decision-making(DM) Deep neural network(DNN) Multi-agent reinforcement learning(MARL) Meta-reinforcement learning(Meta-RL)

Challenges and Opportunities in Preserving Key Structural Features of 3D-Printed Metal/Covalent Organic Framework

在线全文

学校读者我要写书评

暂无评论

Nano-Micro Letters 2024年第8期16卷 362-381页

作者： Ximeng Liu Dan Zhao John Wang Department of Materials Science and Engineering National University of SingaporeSingapore 117574Singapore Department of Chemical and Biomolecular Engineering National University of SingaporeSingapore 117585Singapore national university of Singapore(Chongqing)Research Institute Chongqing 401123People’s Republic of China

Metal-organic framework(MOF)and covalent organic framework(COF)are a huge group of advanced porous materials exhibiting attractive and tunable microstructural features,such as large surface area,tunable pore size,and functional surfaces,which have significant values in various application *** emerging 3D printing technology further provides MOF and COFs(M/COFs)with higher designability of their macrostructure and demonstrates large achievements in their performance by shaping them into advanced 3D ***,the currently available 3D printing M/COFs strategy faces a major challenge of severe destruction of M/COFs’microstructural features,both during and after 3D *** is envisioned that preserving the microstructure of M/COFs in the 3D-printed monolith will bring a great improvement to the related *** this overview,the 3D-printed M/COFs are categorized into M/COF-mixed monoliths and M/COF-covered *** differences in the properties,applications,and current research states are *** up-to-date advancements in paste/scaffold composition and printing/covering methods to preserve the superior M/COF microstructure during 3D printing are further discussed for the two types of 3D-printed M/*** the analysis of the current states of 3D-printed M/COFs,the expected future research direction to achieve a highly preserved microstructure in the 3D monolith is proposed.

关键词： Metal-organic frameworks Covalent organic frameworks 3D printing Microstructure Monolith

A Single-Event-Transient Hardened Phase Locked Loop for Clock and Data Recovery

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第2期33卷 353-361页

作者： Hengzhou YUAN Bin LIANG Hao SANG Weixia XU Yang GUO Xi CHEN national university of Defense Technology

A radiation-hardened phase-locked loop is proposed for phase interpolator clock and data recovery purposes. A sensitive node-compressed charge pump and multi-node cross coupling voltage-controlled oscillators are proposed in this phase-locked loop with the goal of achieving good jitter performance and improving anti-SET（SET,single-event transient） capability. The root mean square（RMS） jitter of the phase-locked loop is reduced from 3.7 ps to 2.58 ps at 2 GHz, while the laser threshold is improved from 120 p J to 370 p J compared to the unhardened phaselocked loop. The hardened phase-locked loop also does not lose its lock state from linear energy transfers（LETs） of 3.3 to 37.3 Me V·cm2/mg.

关键词： Charge pumps Voltage-controlled oscillators Energy exchange Noise Lasers Jitter Phase locked loops

Symmetry quantification and segmentation in STEM imaging through Zernike moments

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 39-48页

作者： Jiadong Dan Cheng Zhang 赵晓续 N.Duane Loh Department of Biological Sciences National University of SingaporeSingapore Center for Bioimaging Sciences(CBIS) National University of SingaporeSingapore Department of Physics National University of SingaporeSingapore School of Materials Science and Engineering Peking UniversityBeijing 100871China

We present a method using Zernike moments for quantifying rotational and reflectional symmetries in scanning transmission electron microscopy(STEM)images,aimed at improving structural analysis of materials at the atomic *** technique is effective against common imaging noises and is potentially suited for low-dose imaging and identifying quantum *** showcase its utility in the unsupervised segmentation of polytypes in a twisted bilayer TaS_(2),enabling accurate differentiation of structural phases and monitoring transitions caused by electron beam *** approach enhances the analysis of structural variations in crystalline materials,marking a notable advancement in the characterization of structures in materials science.

关键词： scanning transmission electron microscopy(STEM) symmetry segmentation