AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Direction of Causationa Learning Procedures
There are procedures now in the machine learning literature that address that very problem. When you're given a two variables, x and y, that are correlated, but you don't o know whether the causal direction goes from x to y or from y to x, these procedures will give you statistical information about way in which the variables are co varying. Now the procedures themselves can actually be tested or calebrated empirically. So there are ways of testing whether you've gotten a success or not. And i think that's an important thing to do. In investigating a causal cognition, it's very important to pay attention to the issue of which are the strategies that make for success.