Score matching for sub-Riemannian bridge sampling
This addresses a challenge in inference for stochastic processes and generative modeling on complex geometric structures, representing an incremental advancement from Riemannian to sub-Riemannian settings.
The paper tackled the problem of simulating conditioned diffusion processes on sub-Riemannian manifolds, where geometry and hypoellipticity complicate score approximation, by modifying machine learning techniques to train score approximators and demonstrating the method on the Heisenberg group with numerical experiments.
Simulation of conditioned diffusion processes is an essential tool in inference for stochastic processes, data imputation, generative modelling, and geometric statistics. Whilst simulating diffusion bridge processes is already difficult on Euclidean spaces, when considering diffusion processes on Riemannian manifolds the geometry brings in further complications. In even higher generality, advancing from Riemannian to sub-Riemannian geometries introduces hypoellipticity, and the possibility of finding appropriate explicit approximations for the score of the diffusion process is removed. We handle these challenges and construct a method for bridge simulation on sub-Riemannian manifolds by demonstrating how recent progress in machine learning can be modified to allow for training of score approximators on sub-Riemannian manifolds. Since gradients dependent on the horizontal distribution, we generalise the usual notion of denoising loss to work with non-holonomic frames using a stochastic Taylor expansion, and we demonstrate the resulting scheme both explicitly on the Heisenberg group and more generally using adapted coordinates. We perform numerical experiments exemplifying samples from the bridge process on the Heisenberg group and the concentration of this process for small time.