标题： 神经网络识别晕白矮星 显示英文标题

标题： Neural Network identification of halo white dwarfs

摘要： 白矮星光度函数已被证明是研究银河系盘的一些特性（如其年龄和局部恒星形成率的过去历史）的优秀工具。 晕中白矮星的观测光度函数的存在可以提供关于其年龄、恒星形成率持续时间的有价值信息，并且还可以限制允许的初始质量函数（IMF）的形状。 然而，主要问题是已经识别为晕星的白矮星数量稀少。 在本信中，我们展示了如何成功地使用一个人工智能算法来分类光谱识别的白矮星种群，使我们能够识别出几个潜在的晕白矮星，并提高其光度函数的显著性。 显示英文摘要

摘要： The white dwarf luminosity function has proven to be an excellent tool to study some properties of the galactic disk such as its age and the past history of the local star formation rate. The existence of an observational luminosity function for halo white dwarfs could provide valuable information about its age, the time that the star formation rate lasted, and could also constrain the shape of the allowed Initial Mass Functions (IMF). However, the main problem is the scarce number of white dwarfs already identified as halo stars. In this Letter we show how an artificial intelligence algorithm can be succesfully used to classify the population of spectroscopically identified white dwarfs allowing us to identify several potential halo white dwarfs and to improve the significance of its luminosity function.