Causal Feature Learning for Utility-Maximizing Agents
This work addresses the challenge of automating high-level causal discovery for researchers in natural and social sciences, but it is incremental as it builds directly on prior CFL methods.
The paper tackles the problem of causal feature learning from low-level data by identifying limitations in the existing CFL method, where it misaligns with pragmatic considerations, and proposes a new technique called PCFL that extends CFL to better handle these cases while maintaining its theoretical properties.
Discovering high-level causal relations from low-level data is an important and challenging problem that comes up frequently in the natural and social sciences. In a series of papers, Chalupka et al. (2015, 2016a, 2016b, 2017) develop a procedure for causal feature learning (CFL) in an effort to automate this task. We argue that CFL does not recommend coarsening in cases where pragmatic considerations rule in favor of it, and recommends coarsening in cases where pragmatic considerations rule against it. We propose a new technique, pragmatic causal feature learning (PCFL), which extends the original CFL algorithm in useful and intuitive ways. We show that PCFL has the same attractive measure-theoretic properties as the original CFL algorithm. We compare the performance of both methods through theoretical analysis and experiments.