Learning Large DAGs by Combining Continuous Optimization and Feedback Arc Set Heuristics
Keywords:Machine Learning (ML)
AbstractBayesian networks represent relations between variables using a directed acyclic graph (DAG). Learning the DAG is an NP-hard problem and exact learning algorithms are feasible only for small sets of variables. We propose two scalable heuristics for learning DAGs in the linear structural equation case. Our methods learn the DAG by alternating between unconstrained gradient descent-based step to optimize an objective function and solving a maximum acyclic subgraph problem to enforce acyclicity. Thanks to this decoupling, our methods scale up beyond thousands of variables.
How to Cite
Gillot, P., & Parviainen, P. (2022). Learning Large DAGs by Combining Continuous Optimization and Feedback Arc Set Heuristics. Proceedings of the AAAI Conference on Artificial Intelligence, 36(6), 6713-6720. https://doi.org/10.1609/aaai.v36i6.20626
AAAI Technical Track on Machine Learning I