Towards Noise-Resilient Quantum Multi-Armed and Stochastic Linear Bandits
This work addresses noise resilience in quantum bandit algorithms, which is crucial for practical applications on current quantum hardware, though it is incremental as it builds on existing quantum methods.
The paper tackles the problem of quantum multi-armed and stochastic linear bandits being sensitive to noise in NISQ devices, and it proposes noise-robust algorithms that improve estimation accuracy and reduce regret while maintaining a quantum advantage over classical methods.
Quantum multi-armed bandits (MAB) and stochastic linear bandits (SLB) have recently attracted significant attention, as their quantum counterparts can achieve quadratic speedups over classical MAB and SLB. However, most existing quantum MAB algorithms assume ideal quantum Monte Carlo (QMC) procedures on noise-free circuits, overlooking the impact of noise in current noisy intermediate-scale quantum (NISQ) devices. In this paper, we study a noise-robust QMC algorithm that improves estimation accuracy when querying quantum reward oracles. Building on this estimator, we propose noise-robust QMAB and QSLB algorithms that enhance performance in noisy environments while preserving the advantage over classical methods. Experiments show that our noise-robust approach improves QMAB estimation accuracy and reduces regret under several quantum noise models.