Towards Efficient Molecular Property Optimization with Graph Energy Based Models

TL;DR

This paper introduces a novel energy-based generative model for efficient chemical property optimization, capable of generating molecules with desired properties without needing explicit labels, and demonstrates superior performance on benchmarks.

Contribution

The authors propose a graph energy-based model that optimizes molecular properties implicitly, advancing the efficiency and robustness of de novo drug design methods.

Findings

Outperforms state-of-the-art methods on chemical benchmarks

Does not require property labels for training

Shows robustness and efficiency in molecule generation

Abstract

Optimizing chemical properties is a challenging task due to the vastness and complexity of chemical space. Here, we present a generative energy-based architecture for implicit chemical property optimization, designed to efficiently generate molecules that satisfy target properties without explicit conditional generation. We use Graph Energy Based Models and a training approach that does not require property labels. We validated our approach on well-established chemical benchmarks, showing superior results to state-of-the-art methods and demonstrating robustness and efficiency towards de novo drug design.