Faster Diffusion Models via Higher-Order Approximation
This provides a training-free acceleration method for diffusion models, which is incremental as it builds on high-order ODE solvers but offers theoretical robustness to inexact scores.
The paper tackles the problem of accelerating diffusion models for sampling from data distributions without retraining, achieving a provable reduction in score function evaluations to order d^{1+2/K} ε^{-1/K} for arbitrary K, applicable to broad distributions without smoothness or log-concavity assumptions.
In this paper, we explore provable acceleration of diffusion models without any additional retraining. Focusing on the task of approximating a target data distribution in $\mathbb{R}^d$ to within $\varepsilon$ total-variation distance, we propose a principled, training-free sampling algorithm that requires only the order of $$ d^{1+2/K} \varepsilon^{-1/K} $$ score function evaluations (up to log factor) in the presence of accurate scores, where $K>0$ is an arbitrary fixed integer. This result applies to a broad class of target data distributions, without the need for assumptions such as smoothness or log-concavity. Our theory is robust vis-a-vis inexact score estimation, degrading gracefully as the score estimation error increases -- without demanding higher-order smoothness on the score estimates as assumed in previous work. The proposed algorithm draws insight from high-order ODE solvers, leveraging high-order Lagrange interpolation and successive refinement to approximate the integral derived from the probability flow ODE. More broadly, our work develops a theoretical framework towards understanding the efficacy of high-order methods for accelerated sampling.