Abstract
Mode-locked fiber lasers make an ideal platform for the investigation of original and complex nonlinear dynamics of ultrashort pulses [1], and provide convenient pulsed sources with scalable repetition rate and energy [2]. To access a wide range of pulsed dynamics from a single setup, nonlinear polarization evolution (NPE) mode locking is the most versatile and popular technique. Indeed, the transfer function of the effective saturable absorber is readily adjusted by the setting of intracavity waveplates, or polarization controllers, which makes the laser explore various dynamics, such as harmonic mode locking [2,3] and bunches of chaotic pulses [4,5]. However, the large number of degrees of freedom, associated with the complex relationship between cavity settings and pulsed regimes, makes the exploration of novel dynamics and the optimization of seeked ones particularly tedious. For these reasons, we propose to use machine learning strategies to achieve and optimize various mode locked states.
© 2015 IEEE
PDF ArticleMore Like This
R. H. Pilny, B. Döpke, C. Brenner, J. C. Balzer, and M. R. Hofmann
CB_P_21 The European Conference on Lasers and Electro-Optics (CLEO/Europe) 2015
Wenjing Zhao, Zhenhong Wang, Zhi Wang, Yange Liu, Guang Yang, and Shangcheng Wang
ASu2A.59 Asia Communications and Photonics Conference (ACP) 2015
R. I. Woodward and E. J. R. Kelleher
STu3P.6 CLEO: Science and Innovations (CLEO:S&I) 2016