GraphDF: A Discrete Flow Model for Molecular Graph Generation
arXiv:2102.01189
Abstract
We consider the problem of molecular graph generation using deep models. While graphs are discrete, most existing methods use continuous latent variables, resulting in inaccurate modeling of discrete graph structures. In this work, we propose GraphDF, a novel discrete latent variable model for molecular graph generation based on normalizing flow methods. GraphDF uses invertible modulo shift transforms to map discrete latent variables to graph nodes and edges. We show that the use of discrete latent variables reduces computational costs and eliminates the negative effect of dequantization. Comprehensive experimental results show that GraphDF outperforms prior methods on random generation, property optimization, and constrained optimization tasks.
arXiv commentAccepted by ICML 2021
References in corpus (14)
- Auto-Encoding Variational Bayes
- Categorical Reparameterization with Gumbel-Softmax
- NICE: Non-linear Independent Components Estimation
- Variational Inference with Normalizing Flows
- Junction Tree Variational Autoencoder for Molecular Graph Generation
- GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
- MolGAN: An implicit generative model for small molecular graphs
- Constrained Graph Variational Autoencoders for Molecule Design
- Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
- GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation
- DIG: A Turnkey Library for Diving into Graph Deep Learning Research
- MolecularRNN: Generating realistic molecular graphs with optimized properties
- Molecular Hypergraph Grammar with its Application to Molecular Optimization
- Graph Residual Flow for Molecular Graph Generation
Cited by in corpus (6)
- Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation
- Artificial Intelligence in Drug Discovery: Applications and Techniques
- SCGG: A Deep Structure-Conditioned Graph Generative Model
- An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design
- Weakly Supervised Concept Map Generation through Task-Guided Graph Translation
- Geometric learning of the conformational dynamics of molecules using dynamic graph neural networks