Search papers, labs, and topics across Lattice.
The paper introduces a semi-supervised training-by-proxy method for generative variational autoencoders (VAEs) to reduce the computational cost of training generative models with expensive multireference simulations. This approach leverages chemical intuition to guide the VAE training process, enabling the generation of novel molecules with desired properties using significantly smaller datasets. The method is demonstrated by generating Dy(III) pentagonal bipyramidal complexes with record values of magnetic anisotropy, starting from only 1k multireference calculations.
Generate novel molecules with targeted properties using generative AI while slashing the need for expensive simulations by 100x.
Generative machine learning models can potentially provide direct access to novel and relevant portions of the full chemical space, overcoming the cost of systematic sampling. However, the training of these models generally requires a large amount of data, often precluding the use of expensive high-level ab initio simulations for this task. The generation of coordination compounds of Dy with large magnetic anisotropy represents a topical example, where multireference simulations of large molecules are necessary to perform reliable predictions. Here, we show that a semi-supervised chemically-inspired training-by-proxy of generative variational autoencoders can reduce the cost associated with building a training set from multireference simulations by two orders of magnitude. We illustrate the power of this approach by generating 100s of new organic ligands for Dy(III) pentagonal bipyramidal complexes exhibiting record values of magnetic anisotropy, while starting from datasets as small as 1k multireference calculations. This work thus paves the way to the computational generation of molecules as complex coordination compounds with target electronic and magnetic properties.
