Learning Kinetic Monte Carlo stochastic dynamics with Deep Generative Adversarial Networks
This work addresses computational efficiency in simulating stochastic dynamics for materials science or physics applications, but it is incremental as it applies a modified GAN to a known bottleneck.
The paper tackled the problem of learning stochastic dynamics for many-particle systems by using a conditional Generative Adversarial Network (GAN) to surrogate traditional Kinetic Monte Carlo simulations, resulting in the network reproducing equilibrium and kinetic properties with deviations of a few percent from exact values.
We show that Generative Adversarial Networks (GANs) may be fruitfully exploited to learn stochastic dynamics, surrogating traditional models while capturing thermal fluctuations. Specifically, we showcase the application to a two-dimensional, many-particle system, focusing on surface-step fluctuations and on the related time-dependent roughness. After the construction of a dataset based on Kinetic Monte Carlo simulations, a conditional GAN is trained to propagate stochastically the state of the system in time, allowing the generation of new sequences with a reduced computational cost. Modifications with respect to standard GANs, which facilitate convergence and increase accuracy, are discussed. The trained network is demonstrated to quantitatively reproduce equilibrium and kinetic properties, including scaling laws, with deviations of a few percent from the exact value. Extrapolation limits and future perspectives are critically discussed.