Efficient Diffusion Training via Min-SNR Weighting Strategy

03/16/2023
by   Tiankai Hang, et al.
0

Denoising diffusion models have been a mainstream approach for image generation, however, training these models often suffers from slow convergence. In this paper, we discovered that the slow convergence is partly due to conflicting optimization directions between timesteps. To address this issue, we treat the diffusion training as a multi-task learning problem, and introduce a simple yet effective approach referred to as Min-SNR-γ. This method adapts loss weights of timesteps based on clamped signal-to-noise ratios, which effectively balances the conflicts among timesteps. Our results demonstrate a significant improvement in converging speed, 3.4× faster than previous weighting strategies. It is also more effective, achieving a new record FID score of 2.06 on the ImageNet 256×256 benchmark using smaller architectures than that employed in previous state-of-the-art.

READ FULL TEXT

page 7

page 15

page 16

page 17

page 18

research
11/20/2021

A Closer Look at Loss Weighting in Multi-Task Learning

Multi-Task Learning (MTL) has achieved great success in various fields, ...
research
11/26/2021

Conditional Image Generation with Score-Based Diffusion Models

Score-based diffusion models have emerged as one of the most promising f...
research
07/01/2023

Residual-based attention and connection to information bottleneck theory in PINNs

Driven by the need for more efficient and seamless integration of physic...
research
11/14/2021

SDnDTI: Self-supervised deep learning-based denoising for diffusion tensor MRI

The noise in diffusion-weighted images (DWIs) decreases the accuracy and...
research
06/01/2023

Addressing Negative Transfer in Diffusion Models

Diffusion-based generative models have achieved remarkable success in va...
research
07/20/2023

Diffusion Sampling with Momentum for Mitigating Divergence Artifacts

Despite the remarkable success of diffusion models in image generation, ...
research
08/08/2022

Analog Bits: Generating Discrete Data using Diffusion Models with Self-Conditioning

We present Bit Diffusion: a simple and generic approach for generating d...

Please sign up or login with your details

Forgot password? Click here to reset