Figure 3 from Molecular relaxation by reverse diffusion with time step prediction | Semantic Scholar (2024)

[PDF] Semantic Reader

High quality image synthesis results are presented using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics, which naturally admit a progressive lossy decompression scheme that can be interpreted as a generalization of autoregressive decoding.

This work proposes the polarizable atom interaction neural network (PaiNN) and improves on common molecule benchmarks over previous networks, while reducing model size and inference time and leverage the equivariant atomwise representations obtained by PaiNN for the prediction of tensorial properties.

This work presents a stochastic differential equation (SDE) that smoothly transforms a complex data distribution to a known prior distribution by slowly injecting noise, and a corresponding reverse-time SDE that transforms the prior distribution back into the data distribution by Slowly removing the noise.

SpookyNet is a deep neural network that explicitly treats electronic degrees of freedom, closing an important remaining gap for models in quantum chemistry.

“Essentials” covers force field and molecular orbital theory, Monte Carlo and Molecular Dynamics simulations, thermodynamic and electronic (spectroscopic) property calculation, condensed phase treatment and a few more topics, and is an alternative to Andrew R. Leach's well-established “Molecular Modeling” and Frank Jensen’s “Introduction to Computational Chemistry”.

This work proposes anautoregressive diffusion model for graph generation that defines a node-absorbing diffusion process that operates directly in the discrete graph space and shows that the two networks can be jointly trained by optimizing a simple lower bound of data likelihood.

A new diffusion model called MolDiff is proposed which can generate atoms and bonds simultaneously while still maintaining their consistency by explicitly modeling the dependence between their relationships.

...

...