DeepAI AI Chat
Log In Sign Up

Learning a Generative Model for Validity in Complex Discrete Structures

by   David Janz, et al.
University of Cambridge
The Alan Turing Institute

Deep generative models have been successfully used to learn representations for high-dimensional discrete spaces by representing discrete objects as sequences, for which powerful sequence-based deep models can be employed. Unfortunately, these techniques are significantly hindered by the fact that these generative models often produce invalid sequences: sequences which do not represent any underlying discrete structure. As a step towards solving this problem, we propose to learn a deep recurrent validator model, which can estimate whether a partial sequence can function as the beginning of a full, valid sequence. This model not only discriminates between valid and invalid sequences, but also provides insight as to how individual sequence elements influence the validity of the overall sequence, and the existence of a corresponding discrete object. To learn this model we propose a reinforcement learning approach, where an oracle which can evaluate validity of complete sequences provides a sparse reward signal. We believe this is a key step toward learning generative models that faithfully produce valid sequences which represent discrete objects. We demonstrate its effectiveness in evaluating the validity of Python 3 source code for mathematical expressions, and improving the ability of a variational autoencoder trained on SMILES strings to decode valid molecular structures.


Actively Learning what makes a Discrete Sequence Valid

Deep learning techniques have been hugely successful for traditional sup...

Discrete Object Generation with Reversible Inductive Construction

The success of generative modeling in continuous domains has led to a su...

Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders

Deep generative models have achieved remarkable success in various data ...

Grammar Variational Autoencoder

Deep generative models have been wildly successful at learning coherent ...

Generative Hybrid Representations for Activity Forecasting with No-Regret Learning

Automatically reasoning about future human behaviors is a difficult prob...

Syntax-Directed Variational Autoencoder for Structured Data

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

Goal-directed Generation of Discrete Structures with Conditional Generative Models

Despite recent advances, goal-directed generation of structured discrete...