Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Equivariant Asynchronous Diffusion: An Adaptive Denoising Schedule for Accelerated Molecular Conformation Generation

Junyi An, Chao Qu, Yun-Fei Shi, Zhijian Zhou, Fenglei Cao, Yuan Qi

TL;DR

Equivariant Asynchronous Diffusion is a novel diffusion model that combines the strengths of both approaches: it uses an asynchronous denoising schedule to better capture molecular hierarchy while maintaining a molecule-level horizon.

Abstract

Recent 3D molecular generation methods primarily use asynchronous auto-regressive or synchronous diffusion models. While auto-regressive models build molecules sequentially, they're limited by a short horizon and a discrepancy between training and inference. Conversely, synchronous diffusion models denoise all atoms at once, offering a molecule-level horizon but failing to capture the causal relationships inherent in hierarchical molecular structures. We introduce Equivariant Asynchronous Diffusion (EAD) to overcome these limitations. EAD is a novel diffusion model that combines the strengths of both approaches: it uses an asynchronous denoising schedule to better capture molecular hierarchy while maintaining a molecule-level horizon. Since these relationships are often complex, we propose a dynamic scheduling mechanism to adaptively determine the denoising timestep. Experimental results show that EAD achieves state-of-the-art performance in 3D molecular generation.

Equivariant Asynchronous Diffusion: An Adaptive Denoising Schedule for Accelerated Molecular Conformation Generation

TL;DR

Equivariant Asynchronous Diffusion is a novel diffusion model that combines the strengths of both approaches: it uses an asynchronous denoising schedule to better capture molecular hierarchy while maintaining a molecule-level horizon.

Abstract

Recent 3D molecular generation methods primarily use asynchronous auto-regressive or synchronous diffusion models. While auto-regressive models build molecules sequentially, they're limited by a short horizon and a discrepancy between training and inference. Conversely, synchronous diffusion models denoise all atoms at once, offering a molecule-level horizon but failing to capture the causal relationships inherent in hierarchical molecular structures. We introduce Equivariant Asynchronous Diffusion (EAD) to overcome these limitations. EAD is a novel diffusion model that combines the strengths of both approaches: it uses an asynchronous denoising schedule to better capture molecular hierarchy while maintaining a molecule-level horizon. Since these relationships are often complex, we propose a dynamic scheduling mechanism to adaptively determine the denoising timestep. Experimental results show that EAD achieves state-of-the-art performance in 3D molecular generation.
Paper Structure (44 sections, 23 equations, 3 figures, 6 tables, 2 algorithms)

This paper contains 44 sections, 23 equations, 3 figures, 6 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Contributions
  3. Preliminaries
  4. Molecular Autoregressive Models
  5. Full-Molecule Diffusion Models
  6. Equivariant Asynchronous Diffusion Model
  7. Revisiting the hierarchical structure in molecule
  8. The Diffusion Process
  9. Independent Noise Level
  10. Optimization Objective
  11. Variable-Size Generation
  12. Sampling
  13. Boundary
  14. Related Work
  15. Autoregressive Methods
  16. ...and 29 more sections

Figures (3)

  • Figure 1: Generation Processes Overview. Left: Autoregressive methods generate atoms sequentially, with each new atom's generation conditioned on the previously generated, noise-free atoms. Middle: Full-molecule diffusion models denoise all atoms simultaneously, iteratively refining a sample of noisy atoms until they are all noise-free. Right: Our proposed EAD model combines the strengths of both approaches by using an asynchronous denoising process. This allows atoms to become noise-free sequentially, where the denoising direction for each atom can benefit from the information of atoms that have already reached a lower noise state. Details of their difference can be found in Section \ref{['sec:diffusion']}.
  • Figure 2: Extra samples generated by EAD trained on the QM9 dataset.
  • Figure : Asynchronous Diffusion Training