Generative Invertible Quantum Neural Networks

TL;DR

This paper introduces a quantum-gate algorithm for Quantum Invertible Neural Networks (QINN) and demonstrates its effectiveness in modeling complex collider data, matching larger classical models in performance.

Contribution

The paper presents the first quantum-gate based algorithm for INNs and applies it to real collider data, showing competitive results with smaller quantum models.

Findings

QINN performs comparably to larger classical INNs.

Hybrid QINN effectively learns complex collider data.

QINN's performance varies with different loss functions.

Abstract

Invertible Neural Networks (INN) have become established tools for the simulation and generation of highly complex data. We propose a quantum-gate algorithm for a Quantum Invertible Neural Network (QINN) and apply it to the LHC data of jet-associated production of a Z-boson that decays into leptons, a standard candle process for particle collider precision measurements. We compare the QINN's performance for different loss functions and training scenarios. For this task, we find that a hybrid QINN matches the performance of a significantly larger purely classical INN in learning and generating complex data.