I understand that exchangeability refers to the notion that the order of data in a sequence does not affect the joint distribution of that data. In a sense, the current data we possess is from the same kind of distribution as future data we may wish to predict (i.e., we could exchange the order and it would have the same joint distribution).
This to me feels heavily related to the notion of covariate shift, albeit this does not make an explicit mention to sequences. To my knowledge, covariate shift is the idea the training data is the same as the distribution of the test data.
My question is therefore, how are these two notions related? I.e., is exchangeability a harsher criteria of covariate shift?
