Attention Based Molecule Generation via Hierarchical Variational Autoencoder

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Sivanesan, Divahar
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917598846779392
author Sivanesan, Divahar
author_facet Sivanesan, Divahar
contents Molecule generation is a task made very difficult by the complex ways in which we represent molecules computationally. A common technique used in molecular generative modeling is to use SMILES strings with recurrent neural networks built into variational autoencoders - but these suffer from a myriad of issues: vanishing gradients, long-range forgetting, and invalid molecules. In this work, we show that by combining recurrent neural networks with convolutional networks in a hierarchical manner, we are able to both extract autoregressive information from SMILES strings while maintaining signal and long-range dependencies. This allows for generations with very high validity rates on the order of 95% when reconstructing known molecules. We also observe an average Tanimoto similarity of .6 between test set and reconstructed molecules, which suggests our method is able to map between SMILES strings and their learned representations in a more effective way than prior works using similar methods.
format Preprint
id arxiv_https___arxiv_org_abs_2402_16854
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Attention Based Molecule Generation via Hierarchical Variational Autoencoder
Sivanesan, Divahar
Biomolecules
Machine Learning
Molecule generation is a task made very difficult by the complex ways in which we represent molecules computationally. A common technique used in molecular generative modeling is to use SMILES strings with recurrent neural networks built into variational autoencoders - but these suffer from a myriad of issues: vanishing gradients, long-range forgetting, and invalid molecules. In this work, we show that by combining recurrent neural networks with convolutional networks in a hierarchical manner, we are able to both extract autoregressive information from SMILES strings while maintaining signal and long-range dependencies. This allows for generations with very high validity rates on the order of 95% when reconstructing known molecules. We also observe an average Tanimoto similarity of .6 between test set and reconstructed molecules, which suggests our method is able to map between SMILES strings and their learned representations in a more effective way than prior works using similar methods.
title Attention Based Molecule Generation via Hierarchical Variational Autoencoder
topic Biomolecules
Machine Learning
url https://arxiv.org/abs/2402.16854