ART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep Scheduling
Read the original on arXiv AI →The paper introduces Adaptive Reparameterized Time (ART), a method that adjusts the clock speed of a reparameterized time variable to redistribute computation across a diffusion sampling trajectory, aiming to minimize Euler discretization error. A randomized companion, ART-RL, reformulates ART as a continuous-time reinforcement learning problem with Gaussian policies, establishing a two-directional bridge between deterministic ART optima and optimal Gaussian policies. Experiments show that ART-RL improves FID scores on CIFAR‑10 across various budgets, and the distilled deterministic schedule can transfer to other datasets without additional inference cost.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.