# Molecular Generative Model Based On Adversarially Regularized Autoencoder

arXiv: https://arxiv.org/abs/1912.05617 · HTML: https://mlsanity.com/paper/1912.05617/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Nov 2019.

## Built on by (most cited first, 8 of 8)

- [ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation](https://mlsanity.com/paper/2309.05853/) (2023)
- [Probabilistic Transformer: Modelling Ambiguities and Distributions for RNA Folding and Molecule Design](https://mlsanity.com/paper/2205.13927/) (2022)
- [Decoding Structure-Spectrum Relationships with Physically Organized Latent Spaces](https://mlsanity.com/paper/2301.04724/) (2023)
- [An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design](https://mlsanity.com/paper/2110.05478/) (2021)
- [Generative models and Bayesian inversion using Laplace approximation](https://mlsanity.com/paper/2203.07755/) (2022)
- [LatentGAN Autoencoder: Learning Disentangled Latent Distribution](https://mlsanity.com/paper/2204.02010/) (2022)
- [CardioGenAI: A Machine Learning-Based Framework for Re-Engineering Drugs for Reduced hERG Liability](https://mlsanity.com/paper/2403.07632/) (2024)
- [A Systematic Evaluation of Molecule Generation Models for De Novo Drug Design: From Benchmarks to Practical Insights](https://mlsanity.com/paper/2609.10099/) (2026)

## Abstract (arXiv)

Deep generative models are attracting great attention as a new promising approach for molecular design. All models reported so far are based on either variational autoencoder (VAE) or generative adversarial network (GAN). Here we propose a new type model based on an adversarially regularized autoencoder (ARAE). It basically uses latent variables like VAE, but the distribution of the latent variables is obtained by adversarial training like in GAN. The latter is intended to avoid both inappropriate approximation of posterior distribution in VAE and difficulty in handling discrete variables in GAN. Our benchmark study showed that ARAE indeed outperformed conventional models in terms of validity, uniqueness, and novelty per generated molecule. We also demonstrated successful conditional generation of drug-like molecules with ARAE for both cases of single and multiple properties control. As a potential real-world application, we could generate EGFR inhibitors sharing the scaffolds of known active molecules while satisfying drug-like conditions simultaneously.