Generating $\pi$-Functional Molecules Using STGG+ with Active Learning

TL;DR

This paper introduces STGG+AL, an active learning framework that enhances the generation of novel, out-of-distribution $ ext{π}$-functional molecules with high oscillator strength, outperforming reinforcement learning methods in molecular design.

Contribution

The paper presents a novel active learning approach integrating STGG+ for designing $ ext{π}$-functional molecules, enabling better exploration of chemical space and out-of-distribution properties.

Findings

Successfully generated molecules with high oscillator strength and NIR absorption.

Outperformed existing reinforcement learning methods in molecular generation tasks.

Provided open-source datasets and tools for future research.

Abstract

Generating novel molecules with out-of-distribution properties is a major challenge in molecular discovery. While supervised learning methods generate high-quality molecules similar to those in a dataset, they struggle to generalize to out-of-distribution properties. Reinforcement learning can explore new chemical spaces but often conducts 'reward-hacking' and generates non-synthesizable molecules. In this work, we address this problem by integrating a state-of-the-art supervised learning method, STGG+, in an active learning loop. Our approach iteratively generates, evaluates, and fine-tunes STGG+ to continuously expand its knowledge. We denote this approach STGG+AL. We apply STGG+AL to the design of organic $\pi$-functional materials, specifically two challenging tasks: 1) generating highly absorptive molecules characterized by high oscillator strength and 2) designing absorptive molecules with reasonable oscillator strength in the near-infrared (NIR) range. The generated molecules are validated and rationalized in-silico with time-dependent density functional theory. Our results demonstrate that our method is highly effective in generating novel molecules with high oscillator strength, contrary to existing methods such as reinforcement learning (RL) methods. We open-source our active-learning code along with our Conjugated-xTB dataset containing 2.9 million $\pi$-conjugated molecules and the function for approximating the oscillator strength and absorption wavelength (based on sTDA-xTB).