Bayesian Analysis and Analytic Continuation of Scattering Amplitudes from Lattice QCD

Miguel Salg, Fernando Romero-López, William I. Jay · 2025

Read Bayesian Analysis and Analytic Continuation of Scattering Amplitudes from Lattice QCD by Miguel Salg, Fernando Romero-López, William I. Jay (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-16161.txt.

Introduction

Salg, Romero-López, and Jay present a Bayesian route from lattice QCD data to scattering amplitudes at real and complex energies. Lattice work gives spectra in a finite box, not amplitudes directly. Their method first rebuilds the amplitude on the real energy axis. A Nevanlinna interpolation then extends it into the complex plane. The approach avoids fixing one formula form in advance. Uncertainty from that choice folds into the result itself.

Tests used both toy data and real lattice datasets. Resonant systems from two collaborations served as proving grounds. The method suits multi-hadron scattering, where resonances hide off the real axis. It gives lattice teams a less biased bridge to physics.

Catalog record

CREATORMiguel Salg, Fernando Romero-López, William I. Jay
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE100 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-16161

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Bayesian Analysis and Analytic Continuation of Scattering Amplitudes from Lattice QCD Miguel Salg, Fernando Romero-López, William I. Jay Abstract We present a novel procedure for analyzing the lattice-QCD spectrum via the finite-volume formalism to obtain constraints on multi-hadron scattering amplitudes at both real and complex energies. This approach combines a Bayesian reconstruction of the scattering amplitude on the real axis with Nevanlinna interpolation for analytic continuation to complex-valued energies. The method is non-parametric, inherently accounting for parametrization dependence within the uncertainty. We demonstrate the applicability of this approach using both toy data and real lattice-QCD data in resonant systems from the HadSpec and BaSc collaborations. I Introduction With the discovery of new unstable hadrons at experimental facilities Koppenburg 2021 ; Liu and Mitchell 2023 , the need to describe the hadron spectrum from first-principles quantum chromodynamics (QCD) has become increasingly urgent. A successful approach involves constraining scattering amplitudes from lattice QCD Bulava et al. 2022 by relating the finite-volume energy spectrum of multi-hadron systems in QCD to the scattering amplitude Luscher 1986 . The resulting amplitude can be analytically continued to the complex energy plane to determine the masses, widths, and couplings of hadronic resonances. Despite its success, this method faces challenges, particularly the need to extrapolate from the real axis to complex-valued energies. The standard approach to obtain multi-hadron scattering amplitudes employs the finite-volume formalism, which relates the finite-volume energy spectrum, characterized by a specific set of quantum numbers, to the infinite-volume scattering amplitude. In the two-body case, each energy level imposes a constraint on the scattering amplitude at the corresponding energy Luscher 1986 ; Luscher 1991 ; Rummukainen and Gottlieb 1995 ; Kim et al. 2005 ; Bernard et al. 2008 ; Leskovec and Prelovsek 2012 ; Gockeler et al. 2012 ; Briceno 2014 —see Ref. Briceno et al. 2018a for a review. In practice, multiple energy levels, obtained using a matrix of correlation functions built from several operators, are used as constraints, and simple functional forms are fitted to these data points. For simple functional forms, like the effective-range expansion, the analytic continuation to complex energies is trivial. However, the choice of a particular parametric description introduces model dependence and systematic uncertainties that are hard to quantify. The challenge is particularly keen in certain cases of phenomenological interest, including broad resonances like the \sigma , far away from threshold where the effective-range expansion is poorly convergent, or in multi-channel systems Bulava et al. 2024a ; Bulava et al. 2024b ; Rodas et al. 2023 ; Briceno et al. 2018b ; Prelovsek et al. 2021 ; Briceno et al. 2017 ; Dudek et al. 2014 ; Woss et al. 2019 ; Wilson et al. 2015 . In this work, we present a novel perspective on the extraction of scattering amplitudes from lattice QCD. This approach is inspired by viewing the finite-volume formalism as an inverse problem, namely, numerical analytic continuation from a finite set of points. While a known amplitude can fully reconstruct the spectrum in the elastic region, the available spectrum only provides a finite set of constraints on the amplitude. Fortunately, unlike other inverse problems such as the reconstruction of spectral densities (see Ref. Rothkopf 2022 for a recent review), this problem is not ill-conditioned, which has led to numerous numerical applications in hadron spectroscopy. This perspective allows us to adapt tools commonly used in inverse problems to the analysis of scattering amplitudes. Specifically, our work receives inspiration from two different tools. One is the Bayesian reconstruction of inverse problems using Gaussian processes Horak et al. 2022 ; Del Debbio et al. 2022 ; Pawlowski et al. 2023 ; Horak et al. 2023 ; Del Debbio et al. 2025 ; Candido et al. 2024 . The other is analytic continuation based on Nevanlinna–Pick interpolation Fei et al. 2021a ; Bergamaschi et al. 2023 . The combination of these two approaches allows us to quantify the uncertainty on scattering amplitudes and resonance poles in a controlled way, including the role of model dependence. Our analysis method proceeds in two steps. First, a parameterization-agnostic Bayesian method, inspired by Gaussian processes, constrains the amplitudes for real-valued energies. Second, Nevanlinna--Pick interpolation leverages the known analytic structure of the amplitude in the complex plane to constrain the location of possible resonance poles. We argue that the Bayesian uncertainty of the first step incorporates the parameterization dependence, while techniques from Nevanlinna--Pick interpolation rigorously bound the uncertainty from ‘‘analytically continuing” from a finite set of points, assuming analyticity in the extrapolation region. 1 1 1 Applying Nevanlinna–Pick interpolation requires mapping the problem to the unit disk using conformal maps. For the rigorous bounds to apply, the conformal maps must correctly bound the codomain of the inverse scattering amplitude. Section IV discusses this point in detail. This paper is organized as follows. We start with a brief review of the two-hadron scattering formalism in finite and infinite volume in Section II . This is followed by a detailed account of our proposed non-parametric analysis strategy: in Section III , we discuss how the Bayesian approach proceeds for the analysis at real energies, while in Section IV , we present Nevanlinna–Pick interpolation for the analytic continuation to complex energies. In Section V , a toy example inspired by the \rho resonance is presented. Numerical applications to maximal-isospin \pi\pi scattering, the \sigma resonance, and the \Lambda(1405) resonance are discussed in Section VI , comparing to existing and published lattice-QCD data by the Hadron Spectrum (HadSpec) and the Baryon Scattering (BaSc) collaborations Bulava et al. 2024a ; Rodas et al. 2023 ; Briceno et al. 2018b ; Briceno et al. 2017 . We conclude in Section VII . Appendices A and B collect a few technical results. II Two-hadron scattering In this work, we consider two-body scattering, with both single and multiple channels. This section reviews the relevant formalism both in finite and infinite volume. It also describes the standard fitting approach used to infer infinite-volume scattering observables from lattice-QCD calculations using the finite-volume formalism. II.1 General considerations of two-hadron scattering amplitudes Figure 1: Locations of singularities in the scattering amplitude \mathcal{M}_{\ell} in the complex momentum and energy planes, inspired by Fig. 1 in Ref. Matuschek et al. 2021 . In the following, we assume spinless particles or other cases where J^{P} is uniquely determined by \ell , such as meson-baryon scattering. The central object of interest is the partial-wave projected two-hadron scattering amplitude, \mathcal{M}_{\ell} , where \ell denotes the partial wave. \mathcal{M}_{\ell} is a function of only the center-of-mass energy of the system E=\sqrt{s} . For a specific scattering channel, we suppose that the two scattering particles have masses m_{1} and m_{2} so that \sqrt{s}=\sqrt{m_{1}^{2}+\bm{q}^{2}}+\sqrt{m_{2}^{2}+\bm{q}^{2}}, (1) where the magnitude of the momentum in the center-of-mass frame is q=|\bm{q}| . A standard representation of the scattering amplitude is in terms of the partial-wave projected K matrix \mathcal{K}_{\ell} : \displaystyle\mathcal{M}_{\ell}^{-1}=\mathcal{K}_{\ell}^{-1}-i\rho, (2) where \rho=q/(\eta 16\pi\sqrt{s}) is the phase space, with \eta=1 for identical particles and \eta=1/2 for non-identical particles. Unitarity and causality dictate that \mathcal{K}_{\ell} be a meromorphic function, which is real-valued for real energies above sub-threshold left-hand cuts. In the single-channel case, the K matrix is related to the phase shift \delta_{\ell} , \mathcal{K}^{-1}_{\ell}=\frac{q\cot\delta_{\ell}}{\eta 16\pi\sqrt{s}}. (3) In other

Read the complete text