Implementing Variational Quantum Classifiers for Higgs Boson Detection

The discovery of the Higgs boson at CERN in 2012 marked a pivotal moment in particle physics. Today, we explore how Variational Quantum Classifiers (VQC) can be applied to classify Higgs boson signals from background noise in ATLAS detector data.

Introduction

The Large Hadron Collider generates petabytes of collision data. Distinguishing genuine Higgs boson events from background processes is a computationally intensive classification problem—one where quantum computing may offer advantages.

The VQC Architecture

A Variational Quantum Classifier consists of three main components:

1. Feature Map: Encodes classical data into quantum states
2. Ansatz: Parameterized quantum circuit that learns decision boundaries
3. Measurement: Extracts classical predictions from quantum states

from qiskit import QuantumCircuit
from qiskit.circuit.library import ZZFeatureMap, RealAmplitudes
from qiskit_machine_learning.algorithms import VQC

# Feature map for encoding
feature_map = ZZFeatureMap(feature_dimension=4, reps=2)

# Trainable ansatz
ansatz = RealAmplitudes(num_qubits=4, reps=3)

# Combine into VQC
vqc = VQC(
    feature_map=feature_map,
    ansatz=ansatz,
    optimizer=COBYLA()
)

Results and Analysis

Our benchmarking on the ATLAS Higgs dataset showed promising results, with the VQC achieving competitive accuracy compared to classical methods while requiring fewer training iterations.

Full research paper coming soon.