贝叶斯块:分而治之、MCMC和细胞聚结方法
Bayesian Blocks: Divide and Conquer, MCMC, and Cell Coalescence Approaches
摘要:
在密集型数据(如时间序列、图像和高维过程)中识别局部结构是天文学中的一个重要问题。由于数据通常由非齐次泊松过程生成,因此适当的模型是将数据空间划分为单元的模型,每个单元由齐次(恒定事件率)泊松过程描述。关键是单元格的大小和位置由数据决定,而不是预定义的,甚至不限于均匀间隔。对于一维时间序列,该方法相当于贝叶斯变点检测。本文简述了三种解决多变点问题的方法:(1)单变点分治法,(2)变点个数的最大后验法,(3)单元合并法。最后一种方法从数据的Voronoi镶嵌开始,因此应该很容易推广到高维空间。
英文摘要:
Identification of local structure in intensive data -- such as time series, images, and higher dimensional processes -- is an important problem in astronomy. Since the data are typically generated by an inhomogeneous Poisson process, an appropriate model is one that partitions the data space into cells, each of which is described by a homogeneous (constant event rate) Poisson process. It is key that the sizes and locations of the cells are determined by the data, and are not predefined or even constrained to be evenly spaced. For one-dimensional time series, the method amounts to Bayesian changepoint detection. Three approaches to solving the multiple changepoint problem are sketched, based on: (1) divide and conquer with single changepoints, (2) maximum posterior for the number of changepoints, and (3) cell coalescence. The last method starts from the Voronoi tessellation of the data, and thus should easily generalize to spaces of higher dimension.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。