Input-Output Optics as a Causal Time Series Mapping: A Generative Machine Learning Solution

Abstract

The response of many-body quantum systems to an optical pulse can be extremely challenging to model. Here we explore the use of neural networks, both traditional and generative, to learn and thus simulate the response of such a system from data. The quantum system can be viewed as performing a complex mapping from an input time-series (the optical pulse) to an output time-series (the systems response) which is often also an optical pulse. Using both the transverse and non-integrable Ising models as examples, we show that not only can temporal convolutional networks capture the input/output mapping generated by the system but can also be used to characterize the complexity of the mapping. This measure of complexity is provided by the size of the smallest latent space that is able to accurately model the mapping. We further find that a generative model, in particular a variational auto-encoder, significantly outperforms traditional auto-encoders at learning the complex response of many-body quantum systems. For the example that generated the most complex mapping, the variational auto-encoder produces outputs that have less than 10% error for more than 90% of inputs across our test data.

References in corpus (29)

• An Introduction to Variational Autoencoders
• QuTiP 2: A Python framework for the dynamics of open quantum systems
• QuTiP: An open-source Python framework for the dynamics of open quantum systems
• Dynamics of quantum phase transition: exact solution in quantum Ising model
• Real-time observation of interfering crystal electrons in high-harmonic generation
• Testing whether all eigenstates obey the Eigenstate Thermalization Hypothesis
• High harmonic imaging of ultrafast many-body dynamics in strongly correlated systems
• Symphony on Strong Field Approximation
• Ultrafast modification of Hubbard $U$ in a strongly correlated material: ab initio high-harmonic generation in NiO
• The antiferromagnetic Ising chain in a mixed transverse and longitudinal magnetic field
• The permutation entropy rate equals the metric entropy rate for ergodic information sources and ergodic dynamical systems
• Machine learning applied to single-shot x-ray diagnostics in an XFEL
• Time series compression: a survey
• High-harmonic generation in one-dimensional Mott insulator
• High harmonic spectroscopy of quantum phase transitions in a high-T${}_c$ superconductor
• High-harmonic generation by electric polarization, spin current, and magnetization
• High-harmonic generation in quantum spin systems
• High-harmonic generation in metallic titanium nitride
• Driven Imposters: Controlling Expectations in Many-Body Systems
• Making Distinct Dynamical Systems Appear Spectrally Identical
• Carrier-wave Rabi flopping signatures in high-order harmonic generation for alkali atoms
• Singularity-free quantum tracking control of molecular rotor orientation
• Emergence of a higher energy structure in strong field ionization with inhomogeneous electric fields
• Purifying electron spectra from noisy pulses with machine learning using synthetic Hamilton matrices
• Dynamics of entanglement in the transverse Ising model
• Dispersion Characterization and Pulse Prediction with Machine Learning
• Deep learning and high harmonic generation
• Optical Indistinguishability via Twinning Fields
• Quantum tracking control of the orientation of symmetric top molecules

Cited by in corpus (1)

• Physics-informed time series analysis with Kolmogorov-Arnold Networks under Ehrenfest constraints