Deep Learning for Optical Tweezers

Optical tweezers exploit light--matter interactions to trap particles rangingfrom single atoms to micrometer-sized eukaryotic cells. For this reason,optical tweezers are a ubiquitous tool in physics, biology, and nanotechnology.Recently, the use of deep learning has started to enhance optical tweezers byimproving their design, calibration, and real-time control as well as thetracking and analysis of the trapped objects, often outperforming classicalmethods thanks to the higher computational speed and versatility of deeplearning. Here, we review how deep learning has already remarkably improvedoptical tweezers, while exploring the exciting, new future possibilitiesenabled by this dynamic synergy. Furthermore, we offer guidelines onintegrating deep learning with optical trapping and optical manipulation in areliable and trustworthy way.