Machine Learning for Self-Tuning Optical Systems

The integration of data-driven machine learning strategies with adaptive control are capable of producing efficient and optimal self-tuning algorithms for nonlinear optical systems. We demonstrate the concept on the specific case of an optical fiber laser cavity where the adaptive, multi parameter extremum-seeking control algorithm obtains and maintains high-energy, single-pulse states. The machine learning algorithm, which is based upon a physically realizable objective function that divides the energy output by the fourth moment of the pulse spectrum, characterizes the cavity itself for rapid state identification and improved optimization. The theory developed is demonstrated on a nonlinear polarization rotation (NPR) based laser using waveplate and polarizer angles to achieve optimal passive mode-locking despite large disturbances to the system. The objective function peaks are high-energy mode-locked states that have a safety margin near parameter regimes where mode-locking breaks down or the multi-pulsing instability occurs. The methods demonstrated can be implemented broadly to optical systems, or more generally to any self-tuning complex.