Equivariant Diffusion for Molecule Generation in 3D

Abstract

This work introduces a diffusion model for molecule generation in 3D that is equivariant to Euclidean transformations. Our E(3) Equivariant Diffusion Model (EDM) learns to denoise a diffusion process with an equivariant network that jointly operates on both continuous (atom coordinates) and categorical features (atom types). In addition, we provide a probabilistic analysis which admits likelihood computation of molecules using our model. Experimentally, the proposed method significantly outperforms previous 3D molecular generative methods regarding the quality of generated samples and efficiency at training time.

Accepted at International Conference on Machine Learning (ICML) 2022

Cited by in corpus (9)

• DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
• On the design space between molecular mechanics and machine learning force fields
• Improved decision making with similarity based machine learning: Applications in chemistry
• Geometric-Facilitated Denoising Diffusion Model for 3D Molecule Generation
• Reflection-Equivariant Diffusion for 3D Structure Determination from Isotopologue Rotational Spectra in Natural Abundance
• Unraveling the Potential of Diffusion Models in Small Molecule Generation
• Molecular Dynamics and Machine Learning Unlock Possibilities in Beauty Design -- A Perspective
• Zero Shot Molecular Generation via Similarity Kernels
• Deep Quantum Graph Dreaming: Deciphering Neural Network Insights into Quantum Experiments