Deep molecular dreaming: inverse machine learning for de-novo molecular design and interpretability with surjective representations. Issue 3 (13th July 2021)

• Record Type: Journal Article
• Authors: Shen, Cynthia; Krenn, Mario; Eppel, Sagi; Aspuru-Guzik, Alán
• Is Part Of: Machine learning: science and technology. Volume 2:Issue 3(2021)
• Publication Date: 2021-07-13
• Subjects: inverse design -- DeepDream -- inceptionism -- cheminformatics -- SELFIES -- de-novo molecular design -- deep generative model; 006.31
• DOI: 10.1088/2632-2153/ac09d6
• Languages: English
• ISSNs: 2632-2153

Abstract

Abstract: Computer-based de-novo design of functional molecules is one of the most prominent challenges in cheminformatics today. As a result, generative and evolutionary inverse designs from the field of artificial intelligence have emerged at a rapid pace, with aims to optimize molecules for a particular chemical property. These models 'indirectly' explore the chemical space; by learning latent spaces, policies, and distributions, or by applying mutations on populations of molecules. However, the recent development of the SELFIES (Krenn 2020 Mach. Learn.: Sci. Technol. 1 045024) string representation of molecules, a surjective alternative to SMILES, have made possible other potential techniques. Based on SELFIES, we therefore propose PASITHEA, a direct gradient-based molecule optimization that applies inceptionism (Mordvintsev 2015) techniques from computer vision. PASITHEA exploits the use of gradients by directly reversing the learning process of a neural network, which is trained to predict real-valued chemical properties. Effectively, this forms an inverse regression model, which is capable of generating molecular variants optimized for a certain property. Although our results are preliminary, we observe a shift in distribution of a chosen property during inverse-training, a clear indication of PASITHEA's viability. A striking property of inceptionism is that we can directly probe the model's understanding of the chemical space on which it is trained. We expect that extending PASITHEA to larger datasets, molecules and more complex properties will lead to advances in the design of new functional molecules as well as the interpretation and explanation of machine learning models.