Robust Quantum Control in Games: an Adversarial Learning Approach

TL;DR

This paper introduces a game-theoretic adversarial learning framework for designing robust quantum controls that withstand uncertainties and noise, improving fidelity and robustness in quantum operations.

Contribution

It proposes a novel family of adversarial learning algorithms, a-GRAPE, for robust quantum control, and demonstrates their effectiveness through numerical experiments.

Findings

Adversarial learning can enhance quantum control robustness.

Balance between fidelity and robustness depends on the learning algorithm.

The proposed methods significantly improve control robustness.

Abstract

High-precision operation of quantum computing systems must be robust to uncertainties and noises in the quantum hardware. In this paper, we show that through a game played between the uncertainties (or noises) and the controls, adversarial uncertainty samples can be generated to find highly robust controls through the search for Nash equilibria (NE). We propose a broad family of adversarial learning algorithms, namely a-GRAPE algorithms, which include two effective learning schemes referred to as the best-response approach and the better-response approach within the game-theoretic terminology, providing options for rapidly learning robust controls. Numerical experiments demonstrate that the balance between fidelity and robustness depends on the details of the chosen adversarial learning algorithm, which can effectively lead to a significant enhancement of control robustness while attaining high fidelity.