Modular Flows: Differential Molecular Generation
This work addresses molecular generation for drug discovery and material synthesis, offering a novel method that improves over incremental approaches by solving specific bottlenecks.
The paper tackled the problem of generating molecules with existing flow models that suffer from issues like artifactual dequantization and lack of permutation invariance, by introducing novel continuous normalizing E(3)-equivariant flows based on graph PDEs, resulting in state-of-the-art performance on QM9 and ZINC250K benchmarks.
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.