Novel Ordering-Based Approaches for Causal Structure Learning in the Presence of Unobserved Variables
Keywords:RU: Causality, ML: Causal Learning, ML: Representation Learning, RU: Graphical Model
AbstractWe propose ordering-based approaches for learning the maximal ancestral graph (MAG) of a structural equation model (SEM) up to its Markov equivalence class (MEC) in the presence of unobserved variables. Existing ordering-based methods in the literature recover a graph through learning a causal order (c-order). We advocate for a novel order called removable order (r-order) as they are advantageous over c-orders for structure learning. This is because r-orders are the minimizers of an appropriately defined optimization problem that could be either solved exactly (using a reinforcement learning approach) or approximately (using a hill-climbing search). Moreover, the r-orders (unlike c-orders) are invariant among all the graphs in a MEC and include c-orders as a subset. Given that set of r-orders is often significantly larger than the set of c-orders, it is easier for the optimization problem to find an r-order instead of a c-order. We evaluate the performance and the scalability of our proposed approaches on both real-world and randomly generated networks.
How to Cite
Mokhtarian, E., Khorasani, M., Etesami, J., & Kiyavash, N. (2023). Novel Ordering-Based Approaches for Causal Structure Learning in the Presence of Unobserved Variables. Proceedings of the AAAI Conference on Artificial Intelligence, 37(10), 12260-12268. https://doi.org/10.1609/aaai.v37i10.26445
AAAI Technical Track on Reasoning Under Uncertainty