Energy-based View of Retrosynthesis

Ruoxi Sun, Hanjun Dai, Li Li, Steven Kearnes, Bo Dai

Keywords: Applications, Retrosynthesis, Energy-based Model

Abstract: Retrosynthesis—the process of identifying a set of reactants to synthesize a target molecule—is of vital importance to material design and drug discovery. Existing machine learning approaches based on language models and graph neural networks have achieved encouraging results. However, the inner connections of these models are rarely discussed, and rigorous evaluations of these models are largely in need. In this paper, we propose a framework that unifies sequence- and graph-based methods as energy-based models (EBMs) with different energy functions. This unified point of view establishes connections between different models and identifies the differences between them, thereby promoting the understanding of model design. We also provide a comprehensive assessment of performance to the community. Moreover, we present a novel “dual” variant within the framework that performs consistent training over Bayesian forward- and backward-prediction by constraining the agreement between the two directions. This model improves the state of the art for template-free approaches where the reaction type is unknown and known.

One-sentence Summary: The paper proposed new energy-based methods for retrosynthesis.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2007.13437/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=IJ-W9WXTGU