Machine learning for prediction of extreme statistics in modulation instability

A central area of research in nonlinear science is the study of instabilities that drive the emergence of extreme events. Unfortunately, experimental techniques for measuring such phenomena often provide only partial characterization. For example, real-time studies of instabilities in nonlinear fibre optics frequently use only spectral data, precluding detailed predictions about the associated temporal properties. Here, we show how Machine Learning can overcome this limitation by predicting statistics for the maximum intensity of temporal peaks in modulation instability based only on spectral measurements. Specifically, we train a neural network based Machine Learning model to correlate spectral and temporal properties of optical fibre modulation instability using data from numerical simulations, and we then use this model to predict the temporal probability distribution based on high-dynamic range spectral data from experiments. These results open novel perspectives in all systems exhibiting chaos and instability where direct time-domain observations are difficult.

[1]  R. I. Woodward,et al.  Towards ‘smart lasers’: self-optimisation of an ultrafast pulse source using a genetic algorithm , 2016, Scientific Reports.

[2]  Ulrich Parlitz,et al.  Observing spatio-temporal dynamics of excitable media using reservoir computing. , 2018, Chaos.

[3]  Ata Mahjoubfar,et al.  Deep Learning in Label-free Cell Classification , 2016, Scientific Reports.

[4]  Steven L. Brunton,et al.  Deep Learning and Model Predictive Control for Self-Tuning Mode-Locked Lasers , 2017, ArXiv.

[5]  David R. Goosman,et al.  Design and field test of a galvanometer deflected streak camera , 2003, International Congress on High-Speed Imaging and Photonics.

[6]  V. Zakharov,et al.  Freak waves as nonlinear stage of Stokes wave modulation instability , 2006 .

[7]  L Larger,et al.  Real time noise and wavelength correlations in octave-spanning supercontinuum generation. , 2013, Optics express.

[8]  P. Grelu,et al.  Real-Time Observation of Internal Motion within Ultrafast Dissipative Optical Soliton Molecules. , 2017, Physical review letters.

[9]  Elias Giacoumidis,et al.  Exceeding the Nonlinear Shannon-Limit in Coherent Optical Communications using 3D Adaptive Machine Learning , 2018 .

[10]  Shie Mannor,et al.  Deep Learning Reconstruction of Ultrashort Pulses , 2018, 2018 Conference on Lasers and Electro-Optics (CLEO).

[11]  Cheng Lei,et al.  The complete optical oscilloscope , 2018 .

[12]  Yibo Zhang,et al.  Phase recovery and holographic image reconstruction using deep learning in neural networks , 2017, Light: Science & Applications.

[13]  Jean-Marc Merolla,et al.  Real-time full-field characterization of transient dissipative soliton dynamics in a mode-locked laser , 2018, Nature Photonics.

[14]  Bahram Jalali,et al.  Fluctuations and correlations in modulation instability , 2012, Nature Photonics.

[15]  Roderick Murray-Smith,et al.  Deep learning for real-time single-pixel video , 2018, Scientific Reports.

[16]  T. Brooke Benjamin,et al.  The disintegration of wave trains on deep water Part 1. Theory , 1967, Journal of Fluid Mechanics.

[17]  Pierre Suret,et al.  Single-shot measurement of phase and amplitude by using a heterodyne time-lens system and ultrafast digital time-holography , 2018 .

[18]  L. Ostrovsky,et al.  Modulation instability: The beginning , 2009 .

[19]  C. M. Reeves,et al.  Function minimization by conjugate gradients , 1964, Comput. J..

[20]  Roberto Morandotti,et al.  Real-time measurements of spontaneous breathers and rogue wave events in optical fibre modulation instability , 2016, Nature Communications.

[21]  John M. Dudley,et al.  Rogue wave early warning through spectral measurements , 2011 .

[22]  J. Dudley,et al.  Supercontinuum generation in photonic crystal fiber , 2006 .

[23]  Adrian Ankiewicz,et al.  Early detection of rogue waves in a chaotic wave field , 2011 .

[24]  B. Jalali,et al.  Time stretch and its applications , 2017, Nature Photonics.

[25]  B. Jalali,et al.  Optical rogue waves , 2007, Nature.

[26]  D. Solli,et al.  Recent progress in investigating optical rogue waves , 2013 .

[27]  Jaideep Pathak,et al.  Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach. , 2018, Physical review letters.

[28]  Cyril Billet,et al.  Emergent rogue wave structures and statistics in spontaneous modulation instability , 2015, Scientific Reports.

[29]  Reza Salem,et al.  Application of space–time duality to ultrahigh-speed optical signal processing , 2013 .

[30]  Michael I. Jordan,et al.  Machine learning: Trends, perspectives, and prospects , 2015, Science.

[31]  L. Larger,et al.  Real-time full bandwidth measurement of spectral noise in supercontinuum generation , 2012, Scientific reports.

[32]  Pierre Suret,et al.  Single-shot observation of optical rogue waves in integrable turbulence using time microscopy , 2016, Nature communications.

[33]  Neil G. R. Broderick,et al.  Observation of soliton explosions in a passively mode-locked fiber laser , 2014, 1409.8373.

[34]  V. I. Bespalov,et al.  Filamentary Structure of Light Beams in Nonlinear Liquids , 1966 .

[35]  K. Goda,et al.  Dispersive Fourier transformation for fast continuous single-shot measurements , 2013, Nature Photonics.

[36]  Miro Erkintalo,et al.  Instabilities, breathers and rogue waves in optics , 2014, Nature Photonics.

[37]  Guohai Situ,et al.  Deep-learning-based ghost imaging , 2017, Scientific Reports.

[38]  Bahram Jalali,et al.  Resolving the build-up of femtosecond mode-locking with single-shot spectroscopy at 90 MHz frame rate , 2016, Nature Photonics.