iGENE: A Differentiable Flux-Tube Gyrokinetic Code in TensorFlow

TL;DR

iGENE is a TensorFlow-based differentiable gyrokinetic simulation tool that enables gradient-based optimization and analysis in plasma turbulence modeling.

Contribution

It introduces a fully-differentiable electromagnetic gyrokinetic code in TensorFlow, facilitating advanced AI-driven plasma research workflows.

Findings

Gradients of simulation outputs can be computed via automatic differentiation.

Gradients are useful for profile predictions despite turbulence stochasticity.

Enables integration of gyrokinetics into AI workflows for optimization and uncertainty quantification.

Abstract

We present iGENE, a fully-differentiable TensorFlow implementation of the electromagnetic local nonlinear gyrokinetic model, which allows us to compute gradients of any simulation output with respect to any input via automatic differentiation. We show that even if the stochastic nature of turbulence prevents the exact evaluation of gradients of nonlinear quantities of interest, they can still be successfully used to perform outer-loop tasks, such as profile predictions. This work enables the integration of gyrokinetics into automated parameter optimization, uncertainty quantification, sensitivity analysis, and AI workflows.