AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Nature of Causal Models
In machine learning, for example, we're often working with very low level features like pixels or like bits in an audio file. In a traditional causal model, you have to build that abstraction directly into the model. But in physical science, it might just be easier to start with whatever the highest resolution of representation is and then focus on the micro interactions between those particles.