Jump to main content
Jump to site search


Constrained Bayesian optimization for automatic chemical design using variational autoencoders

Author affiliations

Abstract

Automatic Chemical Design is a framework for generating novel molecules with optimized properties. The original scheme, featuring Bayesian optimization over the latent space of a variational autoencoder, suffers from the pathology that it tends to produce invalid molecular structures. First, we demonstrate empirically that this pathology arises when the Bayesian optimization scheme queries latent space points far away from the data on which the variational autoencoder has been trained. Secondly, by reformulating the search procedure as a constrained Bayesian optimization problem, we show that the effects of this pathology can be mitigated, yielding marked improvements in the validity of the generated molecules. We posit that constrained Bayesian optimization is a good approach for solving this kind of training set mismatch in many generative tasks involving Bayesian optimization over the latent space of a variational autoencoder.

Graphical abstract: Constrained Bayesian optimization for automatic chemical design using variational autoencoders

Back to tab navigation

Supplementary files

Publication details

The article was received on 12 Aug 2019, accepted on 15 Nov 2019 and first published on 18 Nov 2019


Article type: Edge Article
DOI: 10.1039/C9SC04026A
Chem. Sci., 2020, Advance Article
  • Open access: Creative Commons BY license
    All publication charges for this article have been paid for by the Royal Society of Chemistry

  •   Request permissions

    Constrained Bayesian optimization for automatic chemical design using variational autoencoders

    R. Griffiths and J. M. Hernández-Lobato, Chem. Sci., 2020, Advance Article , DOI: 10.1039/C9SC04026A

    This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Material from this article can be used in other publications provided that the correct acknowledgement is given with the reproduced material.

    Reproduced material should be attributed as follows:

    • For reproduction of material from NJC:
      [Original citation] - Published by The Royal Society of Chemistry (RSC) on behalf of the Centre National de la Recherche Scientifique (CNRS) and the RSC.
    • For reproduction of material from PCCP:
      [Original citation] - Published by the PCCP Owner Societies.
    • For reproduction of material from PPS:
      [Original citation] - Published by The Royal Society of Chemistry (RSC) on behalf of the European Society for Photobiology, the European Photochemistry Association, and RSC.
    • For reproduction of material from all other RSC journals:
      [Original citation] - Published by The Royal Society of Chemistry.

    Information about reproducing material from RSC articles with different licences is available on our Permission Requests page.

Search articles by author

Spotlight

Advertisements