Towards Broad AI for Molecules and Drug Discovery
“Over the last decade, machine learning and Deep Learning methods have paved their way into all kinds of computational task for molecules. The molecular machine learning research community believes that it has made strong progress in * a) activity and property prediction, * b) representation learning and molecular modeling, * c) chemical synthesis and reaction prediction, and * d) generative models for molecules. But have we really made progress? QSAR models have been around since the 1960s and we might have only slightly increased predictive performance. Have these methods deserved the name ”Artificial Intelligence”? In this talk, we provide a perspective recent progress in molecular machine learning, on the essential properties that our AIs should have to make a difference, and steps towards such broad AIs.”
