Artificial intelligence approaches are routinely used in many computer-assisted drug discovery tasks, such as property prediction, de novo molecular design, and retrosynthesis planning. Despite their growing ubiquity, these models are notorious for behaving obscurely, which has generated demand for methods that are more readily accessible to the human mind. In a recent RETHINK White Paper, published in the journal Nature Machine Intelligence, we aim to address this point by summarizing the most promising directions in Explainable Artificial Intelligence (XAI) research, as well as daring a forecast towards future opportunities and potential applications in the context of drug discovery. Jiménez-Luna, J., Grisoni, F. & Schneider, G. (2020) Drug discovery with explainable artificial intelligence. Nature Mach. Intell. 2, 573-584.
Rethinking Design with Artificial Intelligence
