Syntax-Directed Variational Autoencoder for Structured Data

Understand

Deep generative models have been enjoying success in modeling continuous data.

• However it remains challenging to capture the representations for discrete structures with formal grammars and semantics, e.g., computer programs and molecular structures.
• How to generate both syntactically and semantically correct data still remains largely an open problem.
• Inspired by the theory of compiler where the syntax and semantics check is done via syntax-directed translation (SDT), we propose a novel syntax-directed variational autoencoder (SD-VAE) by introducing stochastic lazy attributes.