Score matching for bridges without time-reversals
Paper i proceeding, 2025

We propose a new algorithm for learning bridged diffusion processes using scorematching methods. Our method relies on reversing the dynamics of the forward process and using this to learn a score function, which, via Doob’s h-transform, yields a bridged diffusion process; that is, a process conditioned on an endpoint. In contrast to prior methods, we learn the score term ∇xlogp(t,x;T,y) directly, for given t,y, completely avoiding first learning a timereversal. We compare the performance of our algorithm with existing methods and see that it outperforms using the (learned) time-reversals to learn the score term.

Författare

Elizabeth Louise Baker

Köpenhamns universitet

Moritz Schauer

Göteborgs universitet

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

Stefan Sommer

Köpenhamns universitet

Proceedings of AISTATS 2025

Vol. In press

The 28th International Conference on Artificial Intelligence and Statistics
Mai Koh, Thailand,

Stochastic Continuous-Depth Neural Networks

CHAIR, 2020-08-15 -- .

Styrkeområden

Informations- och kommunikationsteknik

Ämneskategorier (SSIF 2025)

Sannolikhetsteori och statistik

Artificiell intelligens

Relaterade dataset

Code [dataset]

URI: https://github.com/ libbylbaker/forward_bridge.

Mer information

Senast uppdaterat

2025-04-08