Towards AI-assisted Neutrino Flavor Theory Design

TL;DR

This paper introduces AMBer, an AI-driven framework using reinforcement learning to automate and optimize the construction of neutrino flavor models, reducing manual effort and exploring new symmetry groups.

Contribution

The authors develop a novel reinforcement learning-based system for automated model building in neutrino physics, capable of exploring unexamined symmetry groups.

Findings

AMBer successfully reproduces known models in well-studied regions.

It extends exploration to a new, previously unexamined symmetry group.

The approach can be generalized to other theoretical physics model-building tasks.

Abstract

Particle physics theories, such as those which explain neutrino flavor mixing, arise from a vast landscape of model-building possibilities. A model's construction typically relies on the intuition of theorists. It also requires considerable effort to identify appropriate symmetry groups, assign field representations, and extract predictions for comparison with experimental data. We develop an Autonomous Model Builder (AMBer), a framework in which a reinforcement learning agent interacts with a streamlined physics software pipeline to search these spaces efficiently. AMBer selects symmetry groups, particle content, and group representation assignments to construct viable models while minimizing the number of free parameters introduced. We validate our approach in well-studied regions of theory space and extend the exploration to a novel, previously unexamined symmetry group. While demonstrated in the context of neutrino flavor theories, this approach of reinforcement learning with physics software feedback may be extended to other theoretical model-building problems in the future.