Automating Experimental Optics with Sample Efficient Machine Learning Methods

Abstract

As free-space optical systems grow in scale and complexity, troubleshooting becomes increasingly time-consuming and, in the case of remote installations, perhaps impractical. An example of a task that is often laborious is the alignment of a high-finesse optical resonator, which is highly sensitive to the mode of the input beam. In this work, we demonstrate how machine learning can be used to achieve autonomous mode-matching of a free-space optical resonator with minimal supervision. Our approach leverages sample-efficient algorithms to reduce data requirements while maintaining a simple architecture for easy deployment. The reinforcement learning scheme that we have developed shows that automation is feasible even in systems prone to drift in experimental parameters, as may well be the case in real-world applications.

References in corpus (18)

• Photonic quantum technologies
• Satellite-to-ground quantum key distribution
• Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit
• Satellite-to-ground quantum communication using a 50-kg-class micro-satellite
• An optical neural network using less than 1 photon per multiplication
• Linear Optical Quantum Computing in a Single Spatial Mode
• Quantum-limited measurements of optical signals from a geostationary satellite
• Multiparameter optimisation of a magneto-optical trap using deep learning
• Model-based Reinforcement Learning: A Survey
• Exploring Model-based Planning with Policy Networks
• Reconstruction of a Photonic Qubit State with Reinforcement Learning
• All-optical quantum computing using cubic phase gates
• Machine learner optimization of optical nanofiber-based dipole traps for cold ${}^{87}$Rb atoms
• The Raspberry Pi Auto-aligner: Machine Learning for Automated Alignment of Laser Beams
• Predicting atmospheric turbulence for secure quantum communications in free space
• Robust Predictable Control
• Differential wavefront sensing and control using radio-frequency optical demodulation
• Model-free reinforcement learning with noisy actions for automated experimental control in optics