AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is There a Fairness to Causal Modelling in Machine Learning?
Causal modelling is a way of thinking about the problem of fairness. It's not unique to machine learning, but it can be used in many other ways. In practice, we would expect that every causal model captures all the variables af and actually says exactly which ones will influence an outcome. But i think there are lots of different plausible paths that could take us towards this goal.