Speakers
Description
With the introduction of beam-induced background from the decay of muons in the beam pipe of future muon colliders, new techniques must be implemented to filter out the large volume of data generated by such events. The potential presence of long-lived novel particles, which evade conventional time-based cuts, motivates the development of geometrical differentiation methods at the on-chip level. We investigate a range of machine learning algorithms that process simulated data from a single pixel layer module during hit events, producing classifications based on pixel cluster/profile outputs. Our objective is to develop an efficient, lightweight algorithm suitable for future on-chip implementation, enabling robust signal-background differentiation.
