Natural Gradient Optimization for Optical Quantum Circuits

TL;DR

This paper introduces a natural gradient optimization method tailored for optical quantum circuits, accounting for their complex parameter space, and demonstrates its faster convergence and smoother cost decay compared to traditional methods.

Contribution

The work adapts natural gradient descent to complex-valued parameters in optical quantum circuits and compares its performance to standard gradient methods.

Findings

Natural Gradient converges faster than vanilla gradient descent.

Natural Gradient results in smoother cost function decay.

The method allows for larger learning rates without instability.

Abstract

Optical quantum circuits can be optimized using gradient descent methods, as the gates in a circuit can be parametrized by continuous parameters. However, the parameter space as seen by the cost function is not Euclidean, which means that the Euclidean gradient does not generally point in the direction of steepest ascent. In order to retrieve the steepest ascent direction, in this work we implement Natural Gradient descent in the optical quantum circuit setting, which takes the local metric tensor into account. In particular, we adapt the Natural Gradient approach to a complex-valued parameter space. We then compare the Natural Gradient approach to vanilla gradient descent and to Adam over two state preparation tasks: a single-photon source and a Gottesman-Kitaev-Preskill state source. We observe that the NG approach has a faster convergence (due in part to the possibility of using larger learning rates) and a significantly smoother decay of the cost function throughout the optimization.