>I’ve seen repeatedly that it’s much harder for a ML PhD to learn chemistry than for a chemist to learn ML. Perhaps this is selection bias. Among all the chem

>I’ve seen repeatedly that it’s much harder for a ML PhD to learn chemistry than for a chemist to learn ML. Perhaps this is selection bias. Among all the chemists, the ones who will dabble in ML will likely be the chemists with the highest ML related aptitude. In contrast, a ML expert on a chemist project is more likely not being internally driven to explore it but instead has been assigned the work, which means that there is less bias in selection and thus they have less chemistry aptitude.