Particle identification algorithms based on machine learning for STCF

The Super Tau-Charm Facility (STCF) is the next-generation positron-electron collider in China, designed specifically to explore various physics phenomena in the tau-charm energy region. Particle identification (PID) is a crucial component of physics analysis and is essential for precision physics measurements. The STCF imposes high demands on PID accuracy and efficiency to meet its rigorous standards. Over the past few decades, machine learning (ML) techniques have emerged as one of the dominant methodologies for PID in high-energy physics experiments, consistently delivering superior results. This study presents an advanced PID software based on ML algorithms that is developed for STCF to advance physics research. It includes a comprehensive global PID algorithm based on Boosted Decision Trees (BDT) for charged particles, combining information from all sub-detectors, as well as two algorithms based on deep CNN to discriminate charged hadrons with the raw information of Cherenkov detector and to discriminate neutral particles using calorimeter responses, respectively.