天文学进展2017，Vol.35Issue(3)：304-315,12.DOI:10.3969/j.issn.1000-8349.2017.03.03
人工智能在脉冲星候选体筛选中的应用
Application of Artificial Intelligence in the Selection of Pulsar Candidate
摘要
Abstract
Pulsar searching, aiming at discovering pulsars and transient sources such as Rotation RAdio Transients and Fast Radio Bursts(FRB), is the first step of sciences based on pulsars and transient sources. As more pulsars are discovered in pulsar searches, some special kinds of pulsars, including those having very short spin periods and very eccentric orbits, will also be illuminated. These special pulsars provide extreme physical conditions for us to examine our current understands, which is imperative to the research of the dense matter state equation, interplanetary navigation, the interstellar medium, and the measurement of gravitational waves. Nowadays, one typical pulsar search/survey in radio frequencies (e.g. 1.4 GHz) will bring us millions of candidates to confirm. To check the quality of so many candidates, manual operations are too slow and inefficient. Machine learning and artificial intelligence have greatly developed in recent years, helping us identify high quality pulsar candidates from pulsar search results. This paper will discuss the use of aritificial intelligence in pulsar candidates checking, statistics and analysis in pulsar candidates identification, and the efficiency expectations of using similar methods in identifying candidates from Five-hundred-meter Aperture Spherical radio Telescope.关键词
人工智能/脉冲星/候选体筛选Key words
Artificial Intelligence/pulsar/candidate selection分类
天文与地球科学引用本文复制引用
许余云,李菂,刘志杰,王晨,王培,张蕾,潘之辰..人工智能在脉冲星候选体筛选中的应用[J].天文学进展,2017,35(3):304-315,12.基金项目
研究生创新基金项目(研创201528) （研创201528）
国家自然科学基金(U1631132) （U1631132）
中国科学院国际合作局对外合作重点项目(114A11KYSB20160008) （114A11KYSB20160008）
中国科学院战略性先导科技专项(B类)(XDB23000000) （B类）
