[link]
How can we learn causal relationships that explain data? We can learn from nonstationary distributions. If we experiment with different factorizations of relationships between variables we can observe which ones provide better sample complexity when adapting to distributional shift and therefore are likely to be causal. If we consider the variables A and B we can factor them in two ways: $P(A,B) = P(A)P(BA)$ representing a causal graph like $A\rightarrow B$ $P(A,B) = P(AB)P(B)$ representing a causal graph like $A \leftarrow B$ The idea is if we train a model with one of these structures; when adapting to a new shifted distribution of data it will take longer to adapt if the model does not have the correct inductive bias. For example let's say that the true relationship is $A$=Raining causes $B$=Open Umbrella (and not viceversa). Changing the marginal probability of Raining (say because the weather changed) does not change the mechanism that relates $A$ and $B$ (captured by $P(BA)$), but will have an impact on the marginal $P(B)$. So after this distributional shift the function that modeled $P(BA)$ will not need to change because the relationship is the same. Only the function that modeled $P(A)$ will need to change. Under the incorrect factorization $P(B)P(AB)$, adaptation to the change will be slow because both $P(B)$ and $P(AB)$ need to be modified to account for the change in $P(A)$ (due to Bayes rule). Here a difference in sample complexity can be observed when modeling the joint of the shifted distribution. $B\rightarrow A$ takes longer to adapt: https://i.imgur.com/B9FEmA7.png Here the idea is that sample complexity when adapting to a new distribution of data is a heuristic to inform us which causal graph inductive bias is correct. Experimentally this works and they also observe that when models have more capacity it seems that the difference between the models grows. This summary was written with the help of Yoshua Bengio.
Your comment:
