Algorithms for Causal Reasoning in Probability Trees
This work addresses the need for causal reasoning in AI and ML, particularly for tasks like causal induction, by providing algorithms for probability trees, which have been underutilized despite their clean semantics.
The authors tackled the problem of causal reasoning in probability trees, which can represent context-specific causal dependencies, by presenting algorithms that cover the entire causal hierarchy and operate on arbitrary propositional and causal events, expanding causal reasoning to a general class of discrete stochastic processes.
Probability trees are one of the simplest models of causal generative processes. They possess clean semantics and -- unlike causal Bayesian networks -- they can represent context-specific causal dependencies, which are necessary for e.g. causal induction. Yet, they have received little attention from the AI and ML community. Here we present concrete algorithms for causal reasoning in discrete probability trees that cover the entire causal hierarchy (association, intervention, and counterfactuals), and operate on arbitrary propositional and causal events. Our work expands the domain of causal reasoning to a very general class of discrete stochastic processes.