Powerful, transferable representations for molecules through intelligent task selection in deep multitask networks. Issue 23 (1st June 2020)

• Record Type: Journal Article
• Authors: Fare, Clyde; Turcani, Lukas; Pyzer-Knapp, Edward O.
• Is Part Of: Physical chemistry chemical physics. Volume 22:Issue 23(2020)
• Page Start: 13041
• Page End: 13048
• Publication Date: 2020-06-01
• Subjects: Chemistry, Physical and theoretical -- Periodicals; 541.3
• DOI: 10.1039/d0cp02319a
• Languages: English
• ISSNs: 1463-9076

Abstract

Abstract : We develop and test a framework for selecting appropriate chemical datasets to create molecular representations tailored for specific tasks. Abstract : Chemical representations derived from deep learning are emerging as a powerful tool in areas such as drug discovery and materials innovation. Currently, this methodology has three major limitations – the cost of representation generation, risk of inherited bias, and the requirement for large amounts of data. We propose the use of multi-task learning in tandem with transfer learning to address these limitations directly. In order to avoid introducing unknown bias into multi-task learning through the task selection itself, we calculate task similarity through pairwise task affinity, and use this measure to programmatically select tasks. We test this methodology on several real-world data sets to demonstrate its potential for execution in complex and low-data environments. Finally, we utilise the task similarity to further probe the expressiveness of the learned representation through a comparison to a commonly used cheminformatics fingerprint, and show that the deep representation is able to capture more expressive task-based information.