EAGER: Enhanced sensitivity of Dark Matter Detectors via Machine Learning

EAGER：通过机器学习增强暗物质探测器的灵敏度

• 批准号: 2118158
• 负责人: James Battat
• 金额: $ 5.14万
• 依托单位: Wellesley College
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-07-01 至 2023-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2118158&HistoricalAwards=false
• 关键词: EAGER; Enhanced; sensitivity; Dark; Matter

Multiple astronomical observations have established that about 85% of the matter in the universe is not made of known elementary particles. Deciphering the nature of this so-called Dark Matter (DM) is of fundamental importance to cosmology, astrophysics, and high-energy particle physics. Directional dark matter detectors have access to a smoking-gun signature of dark matter – an order unity asymmetry in the angular distribution of recoils induced by Weakly Interacting Massive Particles (WIMPs). Although the leading limits on WIMP dark matter currently come from non-directional experiments, these experiments are rapidly approaching the solar neutrino floor, where the signal will be dominated by neutrinos from the sun and which will make future advances with those technologies more challenging. Directional detectors, however, can reach below the neutrino floor to constrain WIMP dark matter. This EAGER award will leverage Machine Learning (ML) techniques to further enhance the sensitivity of directional DM experiments. The ML techniques and analyses developed under this award would be broadly useful to experiments that employ gas-based Time Projection Chambers. Broader impacts of this work also include the training of a culturally and socioeconomically diverse set of female undergraduate students at Wellesley College, and the enhancement of the physics curriculum through the integration of particle physics experimentation in both teaching and research laboratories. Wellesley is a women's college traditionally ranked in the top 10 for ethnic diversity among liberal arts colleges. By integrating students at Wellesley in all aspects of this experimental particle physics program, the proposed work will broaden the participation of members of underrepresented groups in physics.This work will be applied to existing DRIFT (Directional Recoil Identification From Tracks) data and will establish an analysis pipeline that can be used with new data from directional experiments. Preliminary work using very basic ML techniques has already shown significant improvement in the nuclear recoil detection efficiency, with associated gains in the sensitivity to the WIMP-nucleon cross section. Under this award, the group will undertake a more complete exploration of the ML landscape to further enhance the sensitivity of DRIFT to dark matter.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

多个天文观测已经确定，宇宙中大约85%的物质不是由已知的基本粒子组成的。破译这种所谓的暗物质（DM）的性质对于宇宙学，天体物理学和高能粒子物理学具有根本的重要性。定向暗物质探测器可以获得暗物质的确凿证据--由弱相互作用大质量粒子（WIMPs）引起的反冲角分布中的一阶不对称性。虽然目前对WIMP暗物质的主要限制来自非定向实验，但这些实验正在迅速接近太阳中微子底，在那里信号将由来自太阳的中微子主导，这将使这些技术的未来发展更具挑战性。然而，定向探测器可以到达中微子底部以下，以限制WIMP暗物质。EAGER奖将利用机器学习（ML）技术进一步提高定向DM实验的灵敏度。根据该奖项开发的ML技术和分析将广泛适用于采用基于气体的时间投影室的实验。这项工作的更广泛影响还包括在韦尔斯利学院培训一批具有文化和社会经济多样性的女本科生，以及通过将粒子物理实验纳入教学和研究实验室来加强物理课程。韦尔斯利是一所女子学院，传统上在文科院校中的种族多样性排名前十。通过整合韦尔斯利学生在这个实验粒子物理学计划的各个方面，拟议的工作将扩大在物理学中代表性不足的群体的成员的参与。这项工作将被应用到现有的DRIFT（从轨道定向反冲识别）数据，并将建立一个分析管道，可以使用来自定向实验的新数据。使用非常基本的ML技术的初步工作已经显示出核反冲探测效率的显着改善，以及对WIMP-核子截面的灵敏度的相关增益。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Improved sensitivity of the DRIFT-IId directional dark matter experiment using machine learning

使用机器学习提高 DRIFT-IId 定向暗物质实验的灵敏度

• DOI: 10.1088/1475-7516/2021/07/014
• 发表时间: 2021
• 期刊: Journal of Cosmology and Astroparticle Physics
• 影响因子: 6.4
• 作者: Battat, J.B.R.;Eldridge, C.;Ezeribe, A.C.;Gaunt, O.P.;Gauvreau, J.-L.;Marcelo Gregorio, R.R.;Habich, E.K.K.;Hall, K.E.;Harton, J.L.;Ingabire, I.
• 通讯作者: Ingabire, I.