Reinforcement Learning for Optimal Control in Quantum Sensing Networks

Description

Quantum sensing networks have established new limits to precision measurements in applications such as gravitational wave detection and atomic clocks. As this technology progresses, the operations among individual sensors become increasingly flexible, mimicking those of quantum computers. While this development paves the way for novel applications, identifying optimal control strategies for such sensing networks becomes increasingly complex. To address this challenge, we propose a solution that utilizes classical machine learning by extending an existing framework for model-aware reinforcement learning (RL) with Bayesian estimation by Belliardo et al. to variational algorithms from quantum computing. This approach enables the exploration of both known and novel sensing patterns. Our preliminary results suggest that this framework has the potential to account for arbitrary field patterns across the sensors.

Period	29 Apr 2025
Event title	Quantum Sensing Linz 2025
Event type	Workshop
Location	Linz, AustriaShow on map
Degree of Recognition	Regional