Search papers, labs, and topics across Lattice.
This paper introduces a novel approach to reconstructing population dynamics using Wasserstein gradient flows (WGF) by employing a residual method that minimizes a non-negative loss function. Unlike traditional JKO schemes, which are inflexible and computationally expensive, the proposed stitching method unifies multiple existing techniques and is robust to large gaps in observational data. The authors demonstrate that their method achieves state-of-the-art performance on trajectory inference benchmarks, highlighting its efficiency and effectiveness in modeling complex dynamics.
A new simulation-free method for reconstructing population dynamics outperforms traditional approaches by seamlessly handling large gaps in data.
Reconstructing population dynamics is a central problem in the physical and data sciences. Often, the dynamics are modeled as a Wasserstein gradient flow (WGF): a curve of distributions driven by an energy functional. Though there are multiple mathematical characterizations of a WGF, the dominant algorithmic approach relies on the Jordan--Kinderlehrer--Otto (JKO) scheme. JKO-based methods are inflexible to time discretisation and require solving costly optimal transport problems. We take a residual approach, enforcing the continuity equations via a non-negative loss function whose minimum is the WGF. Combined with a data-fitting divergence, this gives a single global objective. This perspective unifies several existing methods and leads to a new particle-based method, stitching, that is simulation-free and robust to large gaps between observations. We demonstrate that the stitching method achieves state-of-the-art performance across trajectory inference benchmarks. For code see github.com/BasisResearch/wasserstein-residuals.
