I see, I agree. I guess we had different ideas about what constitutes an explanation. This probably does not satisfy your requirements for an explanation. I am also not sure how to generate such a model. It seems like I have an intuition about that this can not be dangerous, but they don’t really have underlying an understanding of what exactly is going on. It’s probably more of the type where I have observed certain things in the past that were not dangerous and this seems sufficiently like them. But at least to some extend that intuition has compressed the actual observed instances, such that I can’t recall all of them in detail to give you the same opportunity to generate an intuition based on them (given that you would believe I report the instances accurately).
I don’t feel like that explains very much. It doesn’t feel like the kind of model that can tell you what side effects the intervention has.
I see, I agree. I guess we had different ideas about what constitutes an explanation. This probably does not satisfy your requirements for an explanation. I am also not sure how to generate such a model. It seems like I have an intuition about that this can not be dangerous, but they don’t really have underlying an understanding of what exactly is going on. It’s probably more of the type where I have observed certain things in the past that were not dangerous and this seems sufficiently like them. But at least to some extend that intuition has compressed the actual observed instances, such that I can’t recall all of them in detail to give you the same opportunity to generate an intuition based on them (given that you would believe I report the instances accurately).