Learnability of multi-instance multi-label learning

Multi-instance multi-label learning(MIML) is a new machine learning framework where one data object is described by multiple instances and associated with multiple class labels.During the past few years,many MIML algorithms have been developed and many applications have been described.However,there lacks theoretical exploration to the learnability of MIML.In this paper,through proving a generalization bound for multi-instance single-label learner and viewing MIML as a number of multi-instance single-label learning subtasks with the correlation among the labels,we show that the MIML hypothesis class constructed from a multi-instance single-label hypothesis class is PAC-learnable.     

浅析《中国图书馆分类法》第5版有关类目的变化与分类标引

通过《中国图书馆分类法》第4版与第5版的对比,指出了两版有关类目在类目设置和分类标引方面的变化,并通过具体实例加以区别和说明。     

The morphological and molecular analyses of a new South American urban tardigrade offer new insights on the biological meaning of the Macrobiotus hufelandi group of species (Tardigrada: Macrobiotidae)

Worldwide knowledge of tardigrade fauna is still limited, and many areas such as South America are not well studied. The collection of new substrates in Argentinean urban areas provided an opportunity to describe the new tardigrade species Macrobiotus kristenseni sp. nov. This species has been studied with an integrative taxonomic approach, analysing its morphology by light and scanning electron microscopy, and considering two genes (cox1 and 18S rRNA) for DNA barcoding and phylogenetic purposes. The species belongs to the Macrobiotus hufelandi group of species, and it is characterized by egg processes in the shape of elongated cones with truncated and enlarged apexes, and by a high genetic distance with respect to closely related species (cox1: 18.2–33.4%). Morphological and molecular data show that the Macrobiotus hufelandi group of species has to be considered a true biological entity; one of the more widespread tardigrade lineages in continental environments. http://www.zoobank.org/urn:lsid:zoobank.org:pub:2CA51ABA-5570-4AFB-8182-CE499D97B0F1     

考虑壅塞特性的电子膨胀阀流量特性模型

分析了电子膨胀阀内部制冷剂的壅塞流动特性，应用均相流和冷冻流模型描述电子膨胀阀的流量特性，用亚稳态特性系数建立膨胀阀质量流量计算模型，并采用某型号电子膨胀阀的实验数据与所建模型的计算值进行对比.结果表明，该模型的计算值与实测值的偏差在-8.59%~7.56%以内.     

水肥高效利用灌排系统调控面临的问题研究

通过调查我国水肥利用的现状,指出我国农业水利中干旱缺水、非点源污染、水资源利用率低下和灌排系统不配套等问题,提出灌区水肥高效利用需深入研究的热点问题,对探讨水肥高效利用和灌排系统管理理论,指导我国灌排系统精量调控,以最少的水肥投入量获取最佳的灌区产量和经济及生态环境效益,具有重要意义.     

RLC电路中类电磁感应透明现象的实验研究

RLC电路可以用来模拟原子系统中的电磁感应透明(Electromagnetically induced transparency,EIT)现象.在电容耦合RLC电路中,通过改变外部信号源的频率,测量了传输到第一个RLC回路的功率随信号频率的变化关系,观察到了由于经典干涉导致的类EIT现象.用两网孔电路模拟了A型原子系统,研究了耦合强度和失谐对类EIT信号的影响；用三网孔电路模拟了倒Y型和N型原子系统,研究了失谐对类EIT信号的影响.     

陈衡哲《西洋史》的教材特色与全球史观念——兼议现行中学历史教材的编写

相对于我们现在的中学世界史教材,民国时期陈衡哲所著的中学教材《西洋史》更胜一筹。其文风引人入胜,其提倡的用整体的、全球的、强调人类历史纵横发展的眼光来观照世界历史的观念更是难能可贵。欲改变现行中学教材的现状,在转变编撰教材理念的同时,还需要真正集学术与写作功力于一身的＂大家＂亲自来编写。     

TSPN: Term-based semantic peer-to-peer networks

In this paper, we propose Term-based Semantic Peer-to-Peer Networks (TSPN) to achieve semantic search. For each peer, TSPN builds a full text index of its documents. Through the analysis of resources, TSPN obtains series of terms, and distributes these terms into the network. Thus, TSPN can use query terms to locate appropriate peers to perform semantic search. Moreover, unlike the traditional structured P2P networks, TSPN uses the terms, not the peers, as the logical nodes of DHT. This can with-stand the impact of network churn. The experimental results show that TSPN has better performance compared with the existing P2P semantic searching algorithms.     

Nonlinear RCMC method for spaceborne/airborne forward-looking bistatic SAR

In the spaceborne/airborne forward-looking bistatic synthetic aperture radar (SA-FBSAR),due to the system platforms’ remarkable velocity difference and the forward-looking mode,the range cell migration (RCM) not only depends on the target’s twodimensional location,but also varies with the range location nonlinearly.And the nonlinearity is not just the slight deviation from the linear part,but exhibits evident nonlinear departure in the RCM trajectory.If the RCM is not properly corrected,nonlinear image distortions would occur.Based on the RCM model,a modified two-step RCM compensation (RCMC) method for SA-FBSAR is proposed.In this method,firstly the azimuth-dependent RCM is compensated by the scaling Fourier transform and the phase multiplication.And then the range-dependent RCM is removed through interpolation.The effectiveness of the proposed RCMC method is verified by the simulation results of both point scatterers and area targets.