Scientists are exploring the use of machine learning to advance the search for fundamental theories of nature within particle physics. The Standard Model of Particle Physics has been successful in predicting experimental outcomes at the Large Hadron Collider, but it is believed that there is New Physics to be discovered Beyond the Standard Model. Machine learning is a useful tool in discovering new physics signals from existing or soon-to-be obtained data.
Machine learning is a rapidly growing field that combines techniques from statistics, data science, and computing. It enables algorithms to learn from data and make predictions over new data. In particle physics, where there are complex relationships between variables across many dimensions of data, machine learning algorithms could provide better discovery and characterization of signals. Researchers are using machine learning to explore potential dark matter signals, including Weakly-Interacting Massive Particles and particles predicted by a simplified Dark Matter Effective Field Theory.
Supervised machine learning methods have proven to have good discriminative power for classifying signals. Researchers have also demonstrated how machine learning can be used to improve the discovery significance of signals, particularly for discovering Top jets and Beyond the Standard Model effects within the Standard Model Effective Field Theory. The researchers also highlight the use of unsupervised learning for discovering signals that are not specified a priori.
The use of machine learning in particle physics is a significant step towards refining signal processing and discovering new physics beyond the Standard Model. With the aid of machine learning, researchers hope to uncover new theories of nature that could enhance our understanding of matter and energy. These findings underline the potential of machine learning in revolutionizing our approach to scientific research, providing more opportunities for complex analysis and refinement of existing theories.
