Conditional Diffusion Guidance under Hard Constraint: A Stochastic Analysis Approach
This addresses safety-critical applications and rare-event simulation where existing methods lack guarantees, offering a novel approach for reliable conditional generation.
The paper tackles conditional generation in diffusion models under hard constraints, where generated samples must satisfy prescribed events with probability one, by developing a principled guidance framework based on Doob's h-transform and martingale representation, with non-asymptotic guarantees and demonstrated effectiveness in enforcing constraints and generating rare-event samples.
We study conditional generation in diffusion models under hard constraints, where generated samples must satisfy prescribed events with probability one. Such constraints arise naturally in safety-critical applications and in rare-event simulation, where soft or reward-based guidance methods offer no guarantee of constraint satisfaction. Building on a probabilistic interpretation of diffusion models, we develop a principled conditional diffusion guidance framework based on Doob's h-transform, martingale representation and quadratic variation process. Specifically, the resulting guided dynamics augment a pretrained diffusion with an explicit drift correction involving the logarithmic gradient of a conditioning function, without modifying the pretrained score network. Leveraging martingale and quadratic-variation identities, we propose two novel off-policy learning algorithms based on a martingale loss and a martingale-covariation loss to estimate h and its gradient using only trajectories from the pretrained model. We provide non-asymptotic guarantees for the resulting conditional sampler in both total variation and Wasserstein distances, explicitly characterizing the impact of score approximation and guidance estimation errors. Numerical experiments demonstrate the effectiveness of the proposed methods in enforcing hard constraints and generating rare-event samples.