A Foundation Model for Lattice QCD: Learning to Understand the Standard Model and Beyond

Projektbeschreibung

Verbesserte Modelle lenken genaue Erforschung neuer Physik

Gittersimulationen der Quantenchromodynamik sind unerlässlich, um genaue theoretische Vorhersagen zu treffen, anhand derer Experimente zur Erforschung von Lücken im Standardmodell geplant werden. Diese Simulationen sind jedoch sehr rechenaufwändig und ungenau. Das Ziel des ERC-finanzierten Projekts FoundLatt besteht darin, dieses Problem durch die Entwicklung des ersten grundlegenden Maschinenlernmodells für Gitter-Quantenchromodynamik zu lösen. Das Mschinenlernmodell wird Komponenten großer, allgemeiner Modelle mit bestehenden spezialisierten Maschinenlernmethoden kombinieren, die bestimmte Aufgaben in Gitter-Quantenchromodynamik-Simulationen beschleunigen. Ergebnis soll ein Modell sein, das nach einmaligem Training dauerhaft genutzt werden kann und den derzeitigen Modellen überlegen ist, die einen enormen, kontinuierlichen und daher unpraktischen Trainingsaufwand erfordern. Damit werden sich hochpräzise Vorhersagen treffen lassen, die als Leitfaden für Experimente im Bereich der Physik jenseits des Standardmodells dienen können.

Ziel

"Despite describing fundamental physics with spectacular accuracy, the Standard Model of particle physics is known to be incomplete. As experimental programmes probe its boundaries at higher energies and with better precision, it is critical that measurements are faced with fully controlled theory predictions. Lattice simulations of Quantum Chromodynamics (QCD) form a major pillar of this effort. However, even with increasing computing capacity, many lattice QCD calculations will not be possible without new techniques.

FoundLatt aims to solve this challenge by developing the first machine learning (ML) foundation model for lattice QCD. As exemplified by the highly successful ChatGPT, these large, general-purpose models have recently shown great promise in other contexts. Combining methods from these contexts with specialised ML methods that already accelerate specific lattice QCD tasks, I will create a lattice QCD foundation model that performs multiple challenging tasks over a range of physical parameters. The main innovation of this programme is a ""train once, use forever"" methodology, which will involve a centralised investment of effort that supersedes impractical training of non-reusable models.

This objective will be achieved via three complementary work packages (WPs). My team and I will develop and train a foundation model capable of two critical tasks, importance sampling and precise estimation of observables, first for a reduced model of QCD (WP1) and then the full theory of QCD (WP2). We will also develop a method to learn a ""quantum perfect"" action for QCD, then fine-tune the foundation model to accomplish this (WP3). Such an action has zero lattice discretisation artefacts, thereby removing one of the largest sources of systematic uncertainty. The main output of FoundLatt will be an ML model that is shared openly with the lattice QCD community, unlocking state-of-the-art theory predictions for the next generation of new physics searches.
"