Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Modular Flows: Differential Molecular Generation

About

Generating new molecules is fundamental to advancing critical applications such as drug discovery and material synthesis. Flows can generate molecules effectively by inverting the encoding process, however, existing flow models either require artifactual dequantization or specific node/edge orderings, lack desiderata such as permutation invariance, or induce discrepancy between the encoding and the decoding steps that necessitates post hoc validity correction. We circumvent these issues with novel continuous normalizing E(3)-equivariant flows, based on a system of node ODEs coupled as a graph PDE, that repeatedly reconcile locally toward globally aligned densities. Our models can be cast as message-passing temporal networks, and result in superlative performance on the tasks of density estimation and molecular generation. In particular, our generated samples achieve state-of-the-art on both the standard QM9 and ZINC250K benchmarks.

Yogesh Verma, Samuel Kaski, Markus Heinonen, Vikas Garg• 2022

Related benchmarks

TaskDatasetResultRank
Molecular GenerationZINC250K
Uniqueness99.7
68
Property optimizationZINC250k (test)
1st Order Metric0.947
33
Molecular GenerationQM9
Validity99.1
30
Molecular GenerationZINC 250K (train/test)
Uniqueness0.997
12
Molecular GenerationQM9 (train test)
Uniqueness99.5
10
Molecular GenerationZINC250K MOSES (test)
FCD0.495
10
Molecule GenerationZINC250K
Generation Time0.46
9
Molecule GenerationQM9
Generation Time0.16
9
Molecule GenerationQM9
FCD0.401
9
Showing 9 of 9 rows

Other info

Code

Follow for update