Partial wave analysis of $\tau^-\to\pi^-\pi^+\pi^-\nu_\tau$ at Belle

TL;DR

This paper discusses simulation studies for analyzing tau decay into three pions and a neutrino at Belle, aiming to understand resonances and improve measurements using advanced modeling techniques.

Contribution

It introduces a novel simulation approach with neural networks and quasi-model-independent partial-wave analysis for tau decay data.

Findings

Achieved higher signal efficiency than previous analyses.

Neural networks effectively model complex background distributions.

Partial-wave analysis extracts resonance parameters accurately.

Abstract

We present simulation studies in preparation for analyzing $\tau^-\to\pi^-\pi^+\pi^-\nu_\tau$ in data from the Belle experiment at the KEK $\mathrm{e}^+\mathrm{e}^-$ collider. Analyzing this decay can shed light on the $\mathrm{a}_1(1260)$ and $\mathrm{a}_1(1420)$ resonances and yield results that improve measurement of the $\tau$ electric and magnetic dipole moments. We show that we can achieve a higher signal efficiency than previous analyses of the same decay. We also demonstrate that neural networks can model our complicated six-dimensional background distributions and that quasi-model-independent partial-wave analysis can extract resonance masses, widths, and production amplitudes and phases.