Beyond Optimization: Harnessing Quantum Annealer Dynamics for Machine Learning

TL;DR

This paper explores using quantum annealer dynamics to generate feature maps for machine learning, showing that optimal annealing times improve classification accuracy and that the participation ratio correlates with model generalization.

Contribution

It introduces a novel approach to machine learning using quantum annealer dynamics and proposes the participation ratio as a measure of model complexity and generalization.

Findings

Short annealing times improve classification accuracy.

Longer annealing times reduce accuracy but lower sampling costs.

Participation ratio correlates with model generalization.

Abstract

Quantum annealing is typically regarded as a tool for combinatorial optimization, but its coherent dynamics also offer potential for machine learning. We present a model that encodes classical data into an Ising Hamiltonian, evolves it on a quantum annealer, and uses the resulting probability distributions as feature maps for classification. Experiments on the quantum annealer machine with the Digits dataset, together with simulations on MNIST, demonstrate that short annealing times yield higher classification accuracy, while longer times reduce accuracy but lower sampling costs. We introduce the participation ratio as a measure of the effective model size and show its strong correlation with generalization.