Control Consistency Losses for Diffusion Bridges

Simulating the conditioned dynamics of diffusion processes, given their initial and terminal states, is an important but challenging problem in the sciences. The difficulty is particularly pronounced for rare events, for which the unconditioned dynamics rarely reach the terminal state. In this work, we leverage a novel self-consistency property of the conditioned dynamics to learn the diffusion bridge drift in an iterative online manner, and demonstrate promising empirical results in a range of settings.