Unbalanced Diffusion Schrödinger Bridge

Matteo Pariset, Ya-Ping Hsieh, Charlotte Bunne, Andreas Krause, Valentin De Bortoli

Primary Area: probabilistic methods (Bayesian methods, variational inference, sampling, UQ, etc.)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Schrödinger bridges, diffusion, entropic interpolation, unbalanced, finite measures

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: Theory and algorithms of entropic interpolation between marginals with arbitrary mass

Abstract: _Schrödinger bridges_ (SBs) provide an elegant framework for modeling the temporal evolution of populations in physical, chemical, or biological systems. Such natural processes are commonly subject to changes in population size over time due to the emergence of new species or birth and death events. However, existing neural parameterizations of SBs such as _diffusion Schrödinger bridges_ ( DSBs) are restricted to settings in which the endpoints of the stochastic process are both _probability measures_ and assume _conservation of mass_ constraints. To address this limitation, we introduce _unbalanced_ DSBs which model the temporal evolution of marginals with arbitrary finite mass. This is achieved by deriving the time reversal of _stochastic differential equations_ (SDEs) with killing and birth terms. We present two novel algorithmic schemes that comprise a scalable objective function for training unbalanced DSBs and provide a theoretical analysis alongside challenging applications on predicting heterogeneous molecular single-cell responses to various cancer drugs and simulating the emergence and spread of new viral variants.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7443