What is it about?
Causal relations in intricate systems such as smarthomes are hard to discover if all components appear at the same level of description. We propose to aggregate tightly-coupled variables to summarize relations between weakly-coupled sub-systems. Then we apply an improved causality discovery method based on causal Bayesian networks in which non "doable" nodes are indirectly reached by acting upon doable variables that have causal relations with them.
Featured Image
Photo by Taylor Vick on Unsplash
Why is it important?
Makes causal discovery scalable.
Read the Original
This page is a summary of: Improving Causal Learning Scalability and Performance using Aggregates and Interventions, ACM Transactions on Autonomous and Adaptive Systems, July 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3607872.
You can read the full text:
Contributors
The following have contributed to this page
