We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Machine Learning

Title: Causal Graph Discovery from Self and Mutually Exciting Time Series

Abstract: We present a generalized linear structural causal model, coupled with a novel data-adaptive linear regularization, to recover causal directed acyclic graphs (DAGs) from time series. By leveraging a recently developed stochastic monotone Variational Inequality (VI) formulation, we cast the causal discovery problem as a general convex optimization. Furthermore, we develop a non-asymptotic recovery guarantee and quantifiable uncertainty by solving a linear program to establish confidence intervals for a wide range of non-linear monotone link functions. We validate our theoretical results and show the competitive performance of our method via extensive numerical experiments. Most importantly, we demonstrate the effectiveness of our approach in recovering highly interpretable causal DAGs over Sepsis Associated Derangements (SADs) while achieving comparable prediction performance to powerful ``black-box'' models such as XGBoost. Thus, the future adoption of our proposed method to conduct continuous surveillance of high-risk patients by clinicians is much more likely.
Comments: See v2 for a previous workshop paper on Interpretable ML in Healthcare (IMLH) at ICML 2021, titled "Causal Graph Recovery for Sepsis-Associated Derangements via Interpretable Hawkes Networks". Also, see arXiv:2301.11336 for a short version with more experiments of our proposed method to learn "strict" DAGs
Subjects: Machine Learning (cs.LG); Statistics Theory (math.ST); Applications (stat.AP); Methodology (stat.ME)
Cite as: arXiv:2106.02600 [cs.LG]
  (or arXiv:2106.02600v5 [cs.LG] for this version)

Submission history

From: Song Wei [view email]
[v1] Fri, 4 Jun 2021 16:59:24 GMT (5847kb,D)
[v2] Sun, 25 Dec 2022 21:13:31 GMT (16628kb,D)
[v3] Tue, 31 Jan 2023 18:24:34 GMT (23787kb,D)
[v4] Sat, 15 Apr 2023 20:30:21 GMT (23784kb,D)
[v5] Tue, 26 Sep 2023 14:50:26 GMT (23738kb,D)

Link back to: arXiv, form interface, contact.