Improved Noise Schedule for Diffusion Training
Keywords: diffusion model, effcient training, noise schedule, image generation
TL;DR: This study proposes a novel noise schedule for training diffusion models that improves efficiency by increasing the sampling frequency around a logSNR of 0, leading to faster convergence and improved performance on benchmarks like ImageNet.
Abstract: Diffusion models have emerged as the de facto choice for generating high-quality visual content across multiple domains.
However, training a single model to predict noise at multiple levels presents significant challenges, requiring numerous iterations and resulting in substantial computational costs.
Various approaches, such as loss weighting strategy design and architectural refinements, have been introduced to expedite convergence and improve model performance.
In this study, we propose a novel approach to design the noise schedule for enhancing the training of diffusion models. Our key insight is that the importance sampling of the logarithm of the Signal-to-Noise ratio ($\log \text{SNR}$), theoretically equivalent to a modified noise schedule, is particularly beneficial for training efficiency when increasing the sample frequency around $\log \text{SNR}=0$. This strategic sampling allows the model to focus on the critical transition point between signal dominance and noise dominance, potentially leading to more robust and accurate predictions.
We empirically demonstrate the superiority of our noise schedule over the standard cosine schedule.
Furthermore, we highlight the advantages of our noise schedule design on the ImageNet benchmark, showing that the designed schedule consistently benefits different prediction targets.
Our findings contribute to the ongoing efforts to optimize diffusion models, potentially paving the way for more efficient and effective training paradigms in the field of generative AI.
Primary Area: generative models
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 3059
Loading