论文标题
Wasserstein梯度流的来回方法
The back-and-forth method for Wasserstein gradient flows
论文作者
论文摘要
我们提出了一种有效计算Wasserstein梯度流量的方法。我们的方法基于Jacobs和Léger引入的来回方法（BFM）的概括，以解决最佳的运输问题。我们通过解决JKO方案的双重问题来发展梯度流。通常，双重问题比原始问题要好得多。这使我们能够有效地进行大规模模拟，以用于包括奇异和非凸的大型内部能量。
We present a method to efficiently compute Wasserstein gradient flows. Our approach is based on a generalization of the back-and-forth method (BFM) introduced by Jacobs and Léger to solve optimal transport problems. We evolve the gradient flow by solving the dual problem to the JKO scheme. In general, the dual problem is much better behaved than the primal problem. This allows us to efficiently run large-scale simulations for a large class of internal energies including singular and non-convex energies.
