检索结果-南通市图书馆

Trustworthy semi-supervised anomaly detection for online‐tooffline logistics business in merchant identification

CAAI Transactions on Intelligence Technology 2024年第3期9卷 544-556页

作者： Yong Li Shuhang Wang Shijie Xu Jiao yin Business School China University of Political Science and LawBeijingChina JD Logistics BeijingChina Southeast University NanjingChina Institute for Sustainable Industries and Liveable Cities Victoria UniversityMelbourneVictoriaAustralia

The rise of online-to-offline(O2O)e-commerce business has brought tremendous opportunities to the logistics *** the online-to-offline logistics business,it is essential to detect anomaly merchants with fraudulent shipping behaviours,such as sending other merchants'packages for profit with their low *** can help reduce the financial losses of platforms and ensure a healthy *** anomaly detection studies have mainly focused on online fraud behaviour detection,such as fraudulent purchase and comment behaviours in ***,these methods are not suitable for anomaly merchant detection in logistics due to the more complex online and offline operation of package-sending behaviours and the interpretable requirements of offline deployment in ***,a semi-supervised multiview fusion-based Anomaly Detection framework in online-to-offline logistics is proposed,which consists of a basic version SemiDet and an attention-enhanced multi-view fusion *** SemiDet,pair-wise data augmentation is first conducted to promote model robustness and address the challenge of limited labelled anomaly ***,SemiDet calculates the anomaly scoring of each merchant with an auto-encoder *** the multi-relationships among logistics merchants,a multi-view attention fusion-based anomaly detection network is further designed to capture merchants'mutual influences and improve the anomaly merchant detection performance.A post-hoc perturbation-based interpretation model is designed to output the importance of different views and ensure the trustworthiness of end-to-end anomaly *** framework based on an eight-month real-world dataset collected from one of the largest logistics platforms in China is evaluated,involving 6128 merchants and 16 million historical order consignor records in *** results show that the proposed model outperforms other baselines in both AUC-ROC and AUC-PR metrics.

关键词： artificial intelligence techniques data fusion deep learning

来源：

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

CCM-FL:Covert communication mechanisms for federated learning in crowd sensing IoT

引用

digital Communications and Networks 2024年第3期10卷 597-608页

作者： Hongruo Zhang Yifei Zou Haofei yin Dongxiao Yu Xiuzhen Cheng Institute of Intelligent Computing School of Computer Science and TechnologyShandong UniversityQingdao266237China

The past decades have witnessed a wide application of federated learning in crowd sensing,to handle the numerous data collected by the sensors and provide the users with precise and customized ***,how to protect the private information of users in federated learning has become an important research *** with the differential privacy(DP)technique and secure multiparty computation(SMC)strategy,the covert communication mechanism in federated learning is more efficient and energy-saving in training the ma-chine learning *** this paper,we study the covert communication problem for federated learning in crowd sensing Internet-of-Things *** from the previous works about covert communication in federated learning,most of which are considered in a centralized framework and experimental-based,we firstly proposes a centralized covert communication mechanism for federated learning among n learning agents,the time complexity of which is O(log n),approximating to the optimal ***,for the federated learning without parameter server,which is a harder case,we show that solving such a problem is NP-hard and prove the existence of a distributed covert communication mechanism with O(log logΔlog n)times,approximating to the optimal solution.Δis the maximum distance between any pair of learning *** analysis and nu-merical simulations are presented to show the performance of our covert communication *** hope that our covert communication work can shed some light on how to protect the privacy of federated learning in crowd sensing from the view of communications.

关键词： Covert communications Federated learning Crowd sensing SINR model

来源：

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

Transient features of graphitization and nitrogen-vacancy color centers in a diamond fabricated by localization femtosecond laser direct writing

引用

Science China Technological Sciences 2024年第8期67卷 2530-2536页

作者： CUI Lin yin SiYu HU ZiFan WANG Lei State Key Laboratory of Integrated Optoelectronics College of Electronic Science and Engineering Jilin University

Femtosecond laser direct writing provides an efficient approach to fabricating single nitrogen vacancy(NV) color centers with a relatively high yield. different from previously reported NV color centers with a random distribution in a bulk diamond or nanocrystals, this gives an opportunity to study the photophysical properties of single NV color centers with precise numbers and positions. However, ultrafast studies on single NV color centers prepared by localization femtosecond laser direct writing are still rare, especially for the graphitization inside a diamond and its relationship with single NV color centers. Here, we report the broadband transient absorption(TA) spectroscopic features of the graphitization and NV color centers in a diamond fabricated by localization femtosecond laser direct writing at room temperature under 400 nm excitation. In comparison with the graphene oxide film, the bleaching features of the graphitization point array in a diamond are similar to reduced graphene oxide,accompanied by excited state absorption signals from local carbon atom vacancy defects in graphene-like structures induced by laser writing. On the other hand, transient features of laser processing array containing single NV color centers with a yield of～50% are different from those of the graphitization point array. Our findings suggest that for ultrashort pulse processing of diamonds, broadband TA spectral signals are sensitive to the surrounding atomic environment of processing sites, which could be applied to laser writing point defects in other materials used as solid-state single photon sources.

关键词： nitrogen-vacancy color center graphitization diamond femtosecond laser direct writing ultrafast spectroscopy

来源：

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Rapid chemical recycling of waste polyester plastics catalyzed by recyclable catalyst

引用

Green Chemical Engineering 2024年第2期5卷 257-265页

作者： Yu-Ji Luo Jia-yin Sun Zhi Li School of Physical Science and Technology ShanghaiTech UniversityShanghai201210China

Waste plastics are serious environmental threats due to their low degradability and low recycling *** and efficient waste plastics recycling technologies are urgently demanded for a sustainable ***,we report a rapid,closed-loop,and streamlined process to convert polyesters such as poly(ethylene terephthalate)(PET)back to its purified *** trifluoromethanesulfonic acid or metal triflates as the recyclable catalyst,polyesters such as PET can be completely depolymerized by simple carboxylic acids within 1 *** coupling this acidolysis with a subsequent hydrogenolysis process,the consumed carboxylic acid was recovered and the closed-loop of PET depolymerization could be *** catalysts and depolymerization agents are fully recycled while only PET and hydrogen are consumed.

关键词： Chemical recycling Waste polyester plastics Acidolysis Recyclable catalyst

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

A robust viscoelastic surfactant tolerating 20% HCl up to 150℃ for oil well stimulation

引用

Petroleum Science 2024年第3期21卷 1865-1874页

作者： Ji Wang Ning Qi Hong-Yao yin Yu-Jun Feng Polymer Research Institute State Key Laboratory of Polymer Materials EngineeringSichuan UniversityChengdu610065SichuanPR.China Tianfu Yongxing Laboratory New Theory and Technology of CO_(2) Capture Research CenterChengdu610065SichuanPR China School of Petroleum Engineering China University of Petroleum(East China)Qingdao266580ShandongPR China

Viscoelastic surfactants(VES)are often used as viscous diverters in acidizing stimulation to prolong the acid consumption time and maximize zonal coverage of the acid for improving well ***,the ceiling temperature of commercial VES cannot exceed 120℃in practical use because of the poor thermal stability and fragile molecular structure,hindering their implementation in hightemperature oil reservoirs,i.e.,≥150℃.Here we synthesized a novel C22-tailed diamine,N-erucaminopropyl-N,N-dimethylamine(EDPA),and examined comparatively its rheological behavior,assemblies morphology and molecular stability in 20 wt%HCl with a commercial VES,erucyl dimethyl amidopropyl betaine(EDAB).The feasibility of EDPA for acidizing stimulation was assessed by acid etching of carbonate rock with its HCl solution at 150℃.Rheological results showed that the 2.5 wt%EDPA—20 wt%HCl solution maintains stable viscosity of 90 m Pa s at 150℃for 60 min,while that of 2.0 wt%EDAB HCl solution is just 1 m Pa s under identical conditions.1H NMR spectra and cryo-TEM observations revealed that the chemical structure and self-assembled architectures of EDPA remained intact in such context,but the EDAB suffered from degradation due to the hydrolysis of the amide group,accounting for the poor heat-resistance and *** reaction rate of 2.5 wt%EDPA HCl solution with carbonate rock was one order of magnitude lower than that of 20 wt%HCl solution at 150℃,underpinning the potential of EDPA to be used in the high-temperature reservoirs *** work improved the thermal tolerance of VES in highly concentrated HCl solution,paving a feasible way for the acidization of high-temperature reservoir environments(~150℃).

关键词： Viscoelastic surfactants Thermal stability Self-diverting acid Wormlike micelles Acidizing stimulation

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Bifunctional interstitial phosphorous doping strategy boosts platinum-zinc alloy for efficient ammonia oxidation reaction and hydrogen evolution reaction

引用

Nano Research 2024年第3期17卷 1182-1189页

作者： Tianqi Yu Kexin Tan Jia Wu Yongjin Zou Shibin yin Guangxi Key Laboratory of Electrochemical Energy Materials School of Chemistry and Chemical EngineeringGuangxi UniversityNanning 530004China Guangxi Key Laboratory of Information Materials Guilin University of Electronic TechnologyGuilin 541004China

It is still a lack of bifunctional catalysts for ammonia oxidation reaction(AOR)and hydrogen evolution reaction(HER)due to their different reaction *** this work,P is doped into PtZn alloy by calcination with NaH_(2)PO_(2) as P source to induce the lattice tensile strain of Pt and the electronic interaction between P and Zn,which optimizes the AOR and HER activity *** sample with the optimal P content can drive the AOR peak current density of 293.6 mA·mgPt^(-1),which is almost 2.7 times of *** HER,the overpotential at^(-1)0 mA·cm^(-2) is only 23 mV with Tafel slope of 34.1 mV·dec^(-1).Furthermore,only 0.59 V is needed to obtain 50 mA·mgPt^(-1) for ammonia electrolysis under a two-electrode ***,this work shows an ingenious method to design bifunctional catalysts for ammonia electrolysis.

关键词： PtZn alloy phosphorous doping ammonia oxidation reaction hydrogen evolution reaction bifunctional catalyst

来源：

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

SWG: an architecture for sparse weight gradient computation

引用

Science China(Information Sciences) 2024年第2期67卷 302-321页

作者： Weiwei WU Fengbin TU Xiangyu LI Shaojun WEI Shouyi yin School of Integrated Circuits Tsinghua University Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology

On-device training for deep neural networks(DNN) has become a trend due to various user preferences and scenarios. The DNN training process consists of three phases, feedforward(FF), backpropagation(BP), and weight gradient(WG) update. WG takes about one-third of the computation in the whole training process. Current training accelerators usually ignore the special computation property of WG and process it in a way similar to FF/BP. Besides, the extensive data sparsity existing in WG, which brings opportunities to save computation, is not well explored. Nevertheless, exploiting the optimization opportunities would meet three underutilization problems, which are caused by(1) the mismatch between WG data dimensions and hardware parallelism,(2) the full sparsity, i.e., the sparsity of feature map(Fmap),error map(Emap), and gradient, and(3) the workload imbalance resulting from irregular sparsity. In this paper, we propose a specific architecture for sparse weight gradient(SWG) computation. The architecture is designed based on hierarchical unrolling and sparsity-aware(HUSA) dataflow to exploit the optimization opportunities of the special computation property and full data sparsity. In HUSA dataflow, the data dimensions are unrolled hierarchically on the hardware architecture. A valid-data trace(VDT) mechanism is embedded in the dataflow to avoid the underutilization caused by the two-sided input sparsity. The gradient is unrolled in PE to alleviate the underutilization induced by output sparsity while maintaining the data reuse opportunities. Besides, we design an intra-and inter-column balancer(IIBLC) to dynamically tackle the workload imbalance problem resulting from the irregular sparsity. Experimental results show that with HUSA dataflow exploiting the full sparsity, SWG achieves a speedup of 12.23× over state-of-the-art gradient computation architecture, Train Ware. SWG helps to improve the energy efficiency of the state-of-the-art training accelerator LNPU from

关键词： CNN training gradient computation sparsity architecture

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Neurotoxicity of tetramethylammonium ion on larval and juvenile zebrafish:Effects on neurobehaviors and multiple biomarkers

引用

Journal of Environmental Sciences 2024年第9期143卷 138-147页

作者： Ruixin Zhang Rui Wang Jiajun Chang GDaniel Sheng Daqiang yin Key Laboratory of Yangtze River Water Environment Ministry of EducationCollege of Environmental Science and EngineeringTongji UniversityShanghai 200092China Shanghai Institute of Pollution Control and Ecological Security Shanghai 200092China

Tetramethylammonium hydroxide(TMAH)is an important compound that utilized and released by the rapidly expanding semiconductor industry,which could hardly be removed by the conventional wastewater treatment *** a cholinergic agonist,the tetramethylammonium ion(TMA^(+))has been reported to induce toxicity to muscular and respiratory systems of mammals and human,however the toxicity on aquatic biota remains poorly *** investigated the neurotoxic effects of TMA^(+)exposure on zebrafish,based on neurobehavior tests and a series of *** inhibitions on the swimming distance of zebrafish larvae were observed when the exposure level exceeded 50 mg/L,and significant alterations on swimming path angles(straight and deflective movements)occurred even at 10 mg/*** tested neurobehavioral endpoints of zebrafish larvae were significantly positively correlated with reactive oxygen species(ROS)and malondialdehyde(MDA),significantly negatively related with the activities of antioxidant enzymes,but not significantly correlated with the level of acetylcholinesterase(AChE).Such relationship indicates that the observed neurotoxic effects on swimming behavior of zebrafish larvae is mainly driven by oxidative stress,rather than the alterations of *** the highest exposure concentration(200 mg/L),TMA^(+)evoked more severe toxicity on zebrafish juveniles,showing significantly stronger elevation on the MDA activity,and greater inhibitions on the activities of antioxidant enzymes and ACh E,suggesting juveniles were more susceptible to TMA^(+)exposure than larval zebrafish.

关键词： Tetramethylammonium Zebrafish Oxidative stress Acetylcholinesterase Neurobehavior

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Planning method of droplet fusion scheduling based on mixedinteger programming

引用

Science China(Technological Sciences) 2024年第1期67卷 157-171页

作者： XIONG JiaCong CHEN JianKui WANG YiXin yin ZhouPing LI YiQun State Key Laboratory of Intelligent Manufacturing Equipment and Technology School of Mechanical Science and EngineeringHuazhong University of Science and TechnologyWuhan 430074China

The emergence of new display devices,such as organic light-emitting diodes,has brought about numerous advantages,including high material utilization,low cost,and high *** devices are manufactured using inkjet printing and possess the potential to become a key technology for display ***,a challenge in achieving this is the display effect that reveals uneven brightness and darkness,which can be avoided by controlling the volume of ink solution in a pixel to within 5%.Currently,the volume difference among the nozzles of commercial printheads does not meet the requirements for volume uniformity,thus challenging the printing ***,designing a suitable printing method that allows for the fusion of different volumes of ink droplets,ultimately reducing the error of the post fusion process,is *** this study,we propose a print display droplet fusion scheduling method comprising two main ***,we use a dichotomous trust domain algorithm to obtain a feasible range of printhead docking point spacings for different nozzle and pixel panel ***,we model the printing process as a droplet fusion scheduling model based on mixed integer programming,with the optimization objective of achieving intra pixel volume uniformity via ensuring the volume uniformity of ink droplets within all *** verified this method through numerical simulations and printing experiments using 394 pixels per inch(ppi)pixel panels and successfully reduced the volume uniformity error among pixels to within 5%.

关键词： printing display droplet fusion mixed-integer programming dichotomous trust domain

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

ReChoreoNet: Repertoire-based Dance Re-choreography with Music-conditioned Temporal and Style Clues

引用

Machine Intelligence Research 2024年第4期21卷 771-781页

作者： Ho yin Au Jie Chen Junkun Jiang Yike Guo Department of Computer Science Hong Kong Baptist UniversityHong Kong 999077China Department of Computer Science and Engineering The Hong Kong University of Science and EngineeringHong Kong 999077China

To generate dance that temporally and aesthetically matches the music is a challenging problem in three ***,the generated motion should be beats-aligned to the local musical ***,the global aesthetic style should be matched between motion and *** third,the generated motion should be diverse and *** address these challenges,we propose ReChoreoNet,which re-choreographs high-quality dance motion for a given piece of music.A data-driven learning strategy is proposed to efficiently correlate the temporal connections between music and motion in a progressively learned cross-modality embedding *** beats-aligned content motion will be subsequently used as autoregressive context and control signal to control a normalizing-flow model,which transfers the style of a prototype motion to the final generated *** addition,we present an aesthetically labelled music-dance repertoire(MDR)for both efficient learning of the cross-modality embedding,and understanding of the aesthetic connections between music and *** demonstrate that our repertoire-based framework is robustly extensible in both content and *** quantitative and qualitative experiments have been carried out to validate the efficiency of our proposed model.

关键词： Generative model cross-modality learning normalizing flow tempo synchronization style transfer.

来源：

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

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：