Augmented Memory: Capitalizing on Experience Replay to Accelerate De Novo Molecular Design

22 May 2023, Version 3
This content is a preprint and has not undergone peer review at the time of posting.


Sample efficiency is a fundamental challenge in de novo molecular design. Ideally, molecular generative models should learn to satisfy desired objectives under minimal oracle evaluations (computational prediction or wet-lab experiment). This problem becomes more apparent when using oracles that can provide increased predictive accuracy but impose a significant cost. Molecular generative models have shown remarkable sample efficiency when coupled with reinforcement learn- ing, as demonstrated in the Practical Molecular Optimization (PMO) benchmark. Here, we propose a novel algorithm called Augmented Memory that combines data augmentation with experience replay. We show that scores obtained from oracle calls can be reused to update the model multiple times. We compare Augmented Memory to previously proposed algorithms and show significantly enhanced sample efficiency in an exploitation task and a drug discovery case study requiring both exploration and exploitation. Our method achieves a new state-of-the-art in the PMO benchmark which enforces a computational budget, and outperforms the previous best performing method on 19/23 tasks.


Molecular generative model
Reinforcement learning
Deep learning
Drug discovery

Supplementary weblinks


Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.