I would speculate, but don't concretely know, that this is what will happen. I know papers in other fields that were just this; analyzing conditions that succes

I would speculate, but don't concretely know, that this is what will happen. I know papers in other fields that were just this; analyzing conditions that successful and failed experiments were performed in and then using ML to derive optimal conditions.