GFlowCausal: Generative Flow Networks for Causal Discovery
Li, Wenqian, Li, Yinchuan, Zhu, Shengyu, Shao, Yunfeng, Hao, Jianye, Pang, Yan
–arXiv.org Artificial Intelligence
Causal discovery aims to uncover causal structure among a set of variables. Score-based approaches mainly focus on searching for the best Directed Acyclic Graph (DAG) based on a predefined score function. However, most of them are not applicable on a large scale due to the limited searchability. Inspired by the active learning in generative flow networks, we propose a novel approach to learning a DAG from observational data called GFlowCausal. It converts the graph search problem to a generation problem, in which direct edges are added gradually. GFlowCausal aims to learn the best policy to generate high-reward DAGs by sequential actions with probabilities proportional to predefined rewards. We propose a plug-and-play module based on transitive closure to ensure efficient sampling. Theoretical analysis shows that this module could guarantee acyclicity properties effectively and the consistency between final states and fully-connected graphs. We conduct extensive experiments on both synthetic and real datasets, and results show the proposed approach to be superior and also performs well in a large-scale setting.
arXiv.org Artificial Intelligence
Mar-10-2023
- Country:
- Asia
- Singapore (0.04)
- China > Ningxia Hui Autonomous Region
- Yinchuan (0.04)
- Asia
- Genre:
- Research Report
- Promising Solution (0.34)
- New Finding (0.34)
- Research Report
- Technology:
- Information Technology > Artificial Intelligence
- Representation & Reasoning
- Search (1.00)
- Uncertainty > Bayesian Inference (0.68)
- Machine Learning
- Neural Networks (1.00)
- Reinforcement Learning (0.93)
- Performance Analysis > Accuracy (0.67)
- Learning Graphical Models > Directed Networks
- Bayesian Learning (1.00)
- Representation & Reasoning
- Information Technology > Artificial Intelligence