An Adjoint Method for Differentiable Fluid Simulation on Flow Maps

TL;DR

This paper introduces a novel adjoint solver for differentiable fluid simulation that leverages shared flow maps for accurate, long-range gradient computation without storing intermediate states, improving efficiency and precision.

Contribution

It proposes a new adjoint method that directly solves on flow maps, enabling efficient, accurate differentiation of incompressible flows without intermediate storage.

Findings

Low memory usage of 6.53GB at high resolution

High accuracy in vortex dynamics tracking

Enables advanced differentiable simulation tasks

Abstract

This paper presents a novel adjoint solver for differentiable fluid simulation based on bidirectional flow maps. Our key observation is that the forward fluid solver and its corresponding backward, adjoint solver share the same flow map as the forward simulation. In the forward pass, this map transports fluid impulse variables from the initial frame to the current frame to simulate vortical dynamics. In the backward pass, the same map propagates adjoint variables from the current frame back to the initial frame to compute gradients. This shared long-range map allows the accuracy of gradient computation to benefit directly from improvements in flow map construction. Building on this insight, we introduce a novel adjoint solver that solves the adjoint equations directly on the flow map, enabling long-range and accurate differentiation of incompressible flows without differentiating intermediate numerical steps or storing intermediate variables, as required in conventional adjoint methods. To further improve efficiency, we propose a long-short time-sparse flow map representation for evolving adjoint variables. Our approach has low memory usage, requiring only 6.53GB of data at a resolution of $192^3$ while preserving high accuracy in tracking vorticity, enabling new differentiable simulation tasks that require precise identification, prediction, and control of vortex dynamics.