LDMX: A sub-GeV Dark Matter discovery eXperiment

Dark matter is one of the major outstanding questions in modern physics, yet we still do not know what particles make up the dark matter in our Universe. While collider experiments have searched extensively for dark matter at GeV and higher masses, the sub-GeV regime remains comparatively unexplored by accelerator-based experiments. This low-mass range is particularly interesting because well-motivated models of dark matter produced in the early Universe can naturally extend into it. The Light Dark Matter eXperiment (LDMX) is a proposed fixed-target experiment at SLAC that will use the 8 GeV electron beam from LCLS-II to search for sub-GeV dark matter. The experiment takes a different approach from conventional collider searches: it will look for collisions in which an electron loses energy and momentum to an invisible particle, producing a characteristic missing-momentum signature. In this talk, I will introduce the physics motivation and experimental concept of LDMX, discuss the design of its detector, and describe how machine-learning techniques (including boosted decision trees and graph neural networks) can help identify rare dark-matter signals amid challenging backgrounds.