WSBD: Freezing-Based Optimizer for Quantum Neural Networks

TL;DR

WSBD is a novel optimizer for quantum neural networks that employs a dynamic freezing strategy to reduce computational costs and improve convergence, outperforming Adam especially on larger models.

Contribution

Introduces WSBD, a parameter-wise freezing optimizer that enhances QNN training efficiency while maintaining full model capacity.

Findings

WSBD converges 63.9% faster than Adam on ground-state-energy problems.

WSBD reduces the number of forward passes per training step.

Parameter-wise freezing outperforms layer-wise freezing in QNNs.

Abstract

The training of Quantum Neural Networks (QNNs) is hindered by the high computational cost of gradient estimation and the barren plateau problem, where optimization landscapes become intractably flat. To address these challenges, we introduce Weighted Stochastic Block Descent (WSBD), a novel optimizer with a dynamic, parameter-wise freezing strategy. WSBD intelligently focuses computational resources by identifying and temporarily freezing less influential parameters based on a gradient-derived importance score. This approach significantly reduces the number of forward passes required per training step and helps navigate the optimization landscape more effectively. Unlike pruning or layer-wise freezing, WSBD maintains full expressive capacity while adapting throughout training. Our extensive evaluation shows that WSBD converges on average 63.9% faster than Adam for the popular ground-state-energy problem, an advantage that grows with QNN size. We provide a formal convergence proof for WSBD and show that parameter-wise freezing outperforms traditional layer-wise approaches in QNNs. Project page: https://github.com/Damrl-lab/WSBD-Stochastic-Freezing-Optimizer.