Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

On Low Rank Directed Acyclic Graphs and Causal Structure Learning

About

Despite several advances in recent years, learning causal structures represented by directed acyclic graphs (DAGs) remains a challenging task in high dimensional settings when the graphs to be learned are not sparse. In this paper, we propose to exploit a low rank assumption regarding the (weighted) adjacency matrix of a DAG causal model to help address this problem. We utilize existing low rank techniques to adapt causal structure learning methods to take advantage of this assumption and establish several useful results relating interpretable graphical conditions to the low rank assumption. Specifically, we show that the maximum rank is highly related to hubs, suggesting that scale-free networks, which are frequently encountered in practice, tend to be low rank. Our experiments demonstrate the utility of the low rank adaptations for a variety of data models, especially with relatively large and dense graphs. Moreover, with a validation procedure, the adaptations maintain a superior or comparable performance even when graphs are not restricted to be low rank.

Zhuangyan Fang, Shengyu Zhu, Jiji Zhang, Yue Liu, Zhitang Chen, Yangbo He• 2020

Related benchmarks

TaskDatasetResultRank
Gene regulatory network inferenceTF Atlas
Recall0.00e+0
7
Gene perturbation response predictionTF-Atlas (10 held-out interventions)
W2105
6
Gene Regulatory Network Discoveryyeast simulated by SERGIO
Recall0.00e+0
6
Gene Regulatory Network DiscoverySERGIO simulated random DAGs
Recall0.06
6
Acyclic Causal DiscoveryPerturb-CITE-seq (Control)
Average Runtime (min)329
4
Acyclic Causal DiscoveryPerturb-CITE-seq Cocult
Average Runtime (min)314.6
4
Acyclic Causal DiscoveryPerturb-CITE-seq IFN
Average Runtime (minutes)366.4
4
Showing 7 of 7 rows

Other info

Follow for update