End-To-End Latent Variational Diffusion Models for Inverse Problems in High Energy Physics

TL;DR

This paper introduces a novel latent variational diffusion model that effectively solves inverse problems in high-energy physics, specifically reconstructing theoretical quantities from detector data at the LHC with high accuracy.

Contribution

The paper proposes a unified latent diffusion architecture combining latent learning with an end-to-end variational framework for inverse problems in particle physics.

Findings

Achieves over 20 times less distributional distance to the truth compared to non-latent baselines.

Outperforms traditional latent diffusion models by a factor of 3 in accuracy.

Effectively reconstructs global distributions of theoretical kinematic quantities.

Abstract

High-energy collisions at the Large Hadron Collider (LHC) provide valuable insights into open questions in particle physics. However, detector effects must be corrected before measurements can be compared to certain theoretical predictions or measurements from other detectors. Methods to solve this \textit{inverse problem} of mapping detector observations to theoretical quantities of the underlying collision are essential parts of many physics analyses at the LHC. We investigate and compare various generative deep learning methods to approximate this inverse mapping. We introduce a novel unified architecture, termed latent variation diffusion models, which combines the latent learning of cutting-edge generative art approaches with an end-to-end variational framework. We demonstrate the effectiveness of this approach for reconstructing global distributions of theoretical kinematic quantities, as well as for ensuring the adherence of the learned posterior distributions to known physics constraints. Our unified approach achieves a distribution-free distance to the truth of over 20 times less than non-latent state-of-the-art baseline and 3 times less than traditional latent diffusion models.