深度学习在引力波数据处理中的应用初探

截至2018-01-16,LIGO已成功探测引力波事件6次.可以预期,引力波探测事件会越来越多,引力波天文学会很快进入到大数据阶段.深度学习在大数据处理方面近年来得到迅速发展.它在数据处理速度,准确度等方面都表现出极大的优势.深度学习在引力波数据处理中的应用讨论还不多.本文引入此问题,并对其进行初步研究.引力波数据最大的特点是强噪声、弱信号.现行的数据处理方法是利用匹配滤波的方式把引力波信号从强噪声中挖掘出来.同时,匹配滤波方法还可以确定引力波源的性质,定量确定其参数.匹配滤波方法的弱点是计算量巨大.这导致数据处理速度很慢.对于将来的大数据引力波天文学,这更将是一个巨大的隐患.匹配滤波方法的另一个潜在问题是,完备准确的理论波形模板是其工作的前提条件.这个潜在问题的后果是很难找到理论预期之外的引力波信号.深度学习的数据处理方法有可能在这些问题上提供出路.同时,深度学习也会遇到其自身的若干困难和问题.本文将从网络结构、训练数据制备、训练优化、对信号识别的泛化能力、对数据的特征图表示以及对特征数据遮挡的响应等方面来展开讨论.     

Fantastic Beasts and where (not) to find them: Local gravitational energy and energy conservation in general relativity

This paper critically examines energy-momentum conservation and local (differential) notions of gravitational energy in General Relativity (GR). On the one hand, I argue that energy-momentum of matter is indeed locally (differentially) conserved: Physical matter energy-momentum 4-currents possess no genuine sinks/sources. On the other hand, global (integral) energy-momentum conservation is contingent on spacetime symmetries. Local gravitational energy-momentum is found to be a supererogatory notion. Various explicit proposals for local gravitational energy-momentum are investigated and found wanting. Besides pseudotensors, the proposals considered include those of Lorentz and Levi-Civita, Pitts and Baker. It is concluded that the ontological commitment we ought to have towards gravitational energy in GR mimics the natural anti-realism/eliminativism towards apparent forces in Newtonian Mechanics.     

对应点集移动相对性建模于脱机签名鉴定研究

提出并论证了欧氏空间中签名的对应点集具有移动相对性,基于该原理构建了一种新的脱机签名鉴定模型,并研究了签名间对应点集获取的方法。在鉴定方案中,待鉴定的脱机签名图像和参考样本的签名图像可以看作欧氏空间内两个不同坐标系中的图像,两者间能够建立相对应的点集合,而欧氏空间平面图像其对应点集具有确定的约束关系,利用该约束关系建立模型可以反映出对应点集移动相对性的大小,因此可以利用该值的差异进行签名真伪鉴定。该方法对于真签名的数量要求较低,并且不需要伪签名来进行训练,适用于脱机签名鉴定中存在的极少样本集情况。实验采用中、英文两种签名数据库验证了该方法的有效性。     

《特种建筑材料》双语教学的探索与实践

就《特种建筑材料》采用双语教学的教材、课件、教学形式、作业、考试等问题进行了探讨,并对《特种建筑材料》开展双语教学一年多来的成效进行了分析,希望对其他课程的双语教学开设有一定借鉴作用。     

物质场论

文章认为物质是以场的形式存在,并通过辐射时空波作用于其他物质。这种作用包括在文章中讨论的量子叠加原理和相对论中的时空相对性原理。     

黑龙江省松嫩平原南部土地质量地球化学评价

以黑龙江省松嫩平原南部8．15万km^2范围表层土壤为研究对象,利用层次分析法、地统计学和模糊评价法从土地的肥力质量和环境质量进行评价,并利用地理信息系统（GIS）技术进一步分析研究区内地球化学质量的空间分布规律和变异性．研究结果表明,黑龙江省松嫩平原南部综合地球化学质量较好,土地的环境质量很高,土质非常清洁、安全．同...     

对脉冲双星PSR1913＋16引力辐射计算的一个注解

本文在牛顿的万有引力定律与爱因斯坦的狭义相对论之间找到一个理论结合点,使质电类比更具理据,以此对陈永明的脉冲双星PSR1913＋16引力辐射的计算提供一个注解。     

Coincidence and reproducibility in the EHT black hole experiment

This paper discusses some philosophical aspects related to the recent publication of the experimental results of the 2017 black hole experiment, namely the first image of the supermassive black hole at the center of galaxy M87. In this paper I present a philosophical analysis of the 2017 Event Horizon Telescope (EHT) black hole experiment. I first present Hacking's philosophy of experimentation. Hacking gives his taxonomy of elements of laboratory science and distinguishes a list of elements. I show that the EHT experiment conforms to major elements from Hacking's list. I then describe with the help of Galison's Philosophy of the Shadow how the EHT Collaboration created the famous black hole image. Galison outlines three stages for the reconstruction of the black hole image: Socio-Epistemology, Mechanical Objectivity, after which there is an additional Socio-Epistemology stage. I subsequently present my own interpretation of the reconstruction of the black hole image and I discuss model fitting to data. I suggest that the main method used by the EHT Collaboration to assure trust in the results of the EHT experiment is what philosophers call the Argument from Coincidence. I show that using this method for the above purpose is problematic. I present two versions of the Argument from Coincidence: Hacking's Coincidence and Cartwright's Reproducibility by which I analyse the EHT experiment. The same estimation of the mass of the black hole is reproduced in four different procedures. The EHT Collaboration concludes: the value we have converged upon is robust. I analyse the mass measurements of the black hole with the help of Cartwright's notion of robustness. I show that the EHT Collaboration construe Coincidence/Reproducibility as Technological Agnosticism and I contrast this interpretation with van Fraassen's scientific agnosticism.     

Interpreting cosmic no hair theorems: Is fatalism about the far future of expanding cosmological models unavoidable?

It may be tempting to interpret cosmic no hair theorems (establishing that, under certain conditions, a cosmological model with accelerating expansion approaches de Sitter metric in the future) as implying a form of fatalism about a future state of the universe. In this note I explore (rather general) philosophical and (more concrete) physical reasons to resist such temptation. There are good reasons to think that (even locally) futures of expanding cosmological models may be unlike de Sitter.     

宇宙年龄和宇宙物质密度

根据爱因斯坦广义相对论,时空弯曲,宇宙是有限的。现在已被主流物理学家普遍接受的是标准宇宙论,按标准宇宙论经过严格的数学推导,演算得到了宇宙的年龄、宇宙物质密度。作者根据广大天体物理工作者认可的观察数据和结论,通过初等数学的计算,得到的结果大相径庭。根本没天体物工作者寻找的暗物质。