Cai: Transformers to transform Scattering Amplitudes Calculation / Monzani: Anomaly detection for rare event searches / Terao: Hunting a Ghost Particle in Big Images

Nov 09, 2023 - 11:00 am to 12:00 pm

SLAC, Kavli 3rd Floor Conf. Room

Tianji Cai, Maria Elena Monzani, and Kazuhiro Terao (SLAC) In Person and zoom

Zoom Recording

Maria Elena Monzani: Anomaly detection for rare event searches

The LUX-ZEPLIN (LZ) Dark Matter detector is projected to accumulate a massive dataset of many petabytes of data and record several billions of particle interactions, only a handful of which might be produced by dark matter interactions. Identifying such interactions requires leveraging advanced detector design and state-of-the-art machine learning algorithms. The talk will present the application of anomaly detection techniques to the identification of ultra-rare interactions in cosmic frontier experiments.

Kazuhiro Terao: Hunting a Ghost Particle in Big Images

Machine learning methods have made many impacts in experimental high energy physics. They can accelerate our workflows and push the limit of physics sensitivity. One of the major challenges in the U.S. neutrino experiments are efficient and precise identification of neutrino signal in a big image data. Powerful deep learning algorithms have been incorporated into an end-to-end optimizable data reconstruction chain to address those challenges. In this talk, I will discuss how they are making impacts in our experiment, new challenges, and the directions for the next generation of data reconstruction methods that may overcome issues we face today.

Tianji Cai: Transformers to transform Scattering Amplitudes Calculation

AI for fundamental physics is now a burgeoning field, with numerous efforts pushing the boundaries of experimental and theoretical physics, as well as machine learning research itself. In this talk, I will introduce a recent innovative application of Natural Language Processing to state-of-the-art calculations in theoretical particle physics. Specifically, we use Transformers to predict complicated mathematical expressions that represent scattering amplitudes in planar N=4 Yang-Mills theory—a quantum field theory closely related to the real-world QCD that describes Higgs boson production at the Large Hadron Collider. Our first results have demonstrated great promises of Transformers for amplitude calculations, opening the door for an exciting new scientific paradigm where discoveries and human insights are inspired and aided by an AI agent.