Machine learning techniques for detecting topological avatars of new physics.

Area of Science:

• High-energy physics
• Particle physics
• Data science

Background:

• Traditional searches for new physics particles in nuclear track detectors (NTDs) are manual and time-consuming.
• Identifying anomalies in NTDs could indicate physics beyond the standard model.
• Automated image acquisition and data science offer new possibilities.

Purpose of the Study:

• To explore the potential of modern data science, including machine learning, for analyzing NTD data.
• To accelerate the search for anomalies indicative of new physics.
• To discuss applications within the MoEDAL experiment at CERN.

Main Methods:

• Utilizing automated image acquisition for NTD data.
• Applying machine learning algorithms for data processing and analysis.
• Investigating anomaly detection techniques.

Main Results:

• Machine learning can significantly accelerate the identification of regions of interest in NTDs.
• Data science methods offer a powerful tool for discovering subtle anomalies.
• The approach is being explored in the context of the MoEDAL experiment.

Conclusions:

• Modern data science and machine learning can revolutionize the search for new physics.
• Automated analysis of NTDs holds significant promise for discovering physics beyond the standard model.
• The MoEDAL experiment can benefit from these advanced analytical techniques.