Networks of Causal Abstractions: A Sheaf-theoretic Framework

ArXi:2509.25236v3 Announce Type: replace A core challenge in causal artificial intelligence is the principled coordination of multiple, imperfect, and subjective causal perspectives arising from distributed agents with limited and heterogeneous access to the environment. This problem has received little formal treatment, as the existing framework assumes a single shared global causal model. This work At the theoretical level, we provide a categorical formulation of MCMs and characterize key properties of CANs, including consistency and smoothness.