基于条件独立性检验的非稳态因果发现方法

doi:10.3778/j.issn.1002-8331.2301-0083

计算机工程与应用 ›› 2024, Vol. 60 ›› Issue (10): 113-120.DOI: 10.3778/j.issn.1002-8331.2301-0083

基于条件独立性检验的非稳态因果发现方法

郝志峰，张维杰，蔡瑞初，陈薇

1.广东工业大学计算机学院，广州 510006
2.汕头大学理学院，广东汕头 515063

出版日期:2024-05-15 发布日期:2024-05-15

Non-Stationary Causal Discovery Method Based on Conditional Independence Test

HAO Zhifeng, ZHANG Weijie, CAI Ruichu, CHEN Wei

1.School of Computer Science, Guangdong University of Technology, Guangzhou 510006, China
2.College of Science, Shantou University, Shantou, Guangdong 515063, China

Online:2024-05-15 Published:2024-05-15

摘要/Abstract

摘要： 非稳态时间序列数据之间的因果关系发现是非常重要但极具挑战的问题。现有的工作主要假设观察数据随着时间或领域发生变化。上述假设使得相关方法需要引入时间或领域作为先验知识，无法应用于分段稳态的非稳态场景。因此，提出了一种基于条件独立性检验的非稳态因果关系发现算法。首先使用变化点检测方法来识别非稳态变化的时间点，然后将上一步的时间点进行区间划分，用基于条件独立性检验的时序因果关系发现算法推断局部稳态因果结构。在仿真和真实世界数据上的实验证明了该方法的有效性。

关键词: 因果关系发现, 非稳态, 因果网络

Abstract: Causal discovery on non-stationary time-series data is of great importance but challenging. Existing works mainly assume that the observed data change with time or domain, which requires the introduction of time or domain as prior knowledge. The aforementioned methods are usually unavailable on the segmented-stationary non-stationary scenarios. Therefore, this paper proposes a non-stationary causal discovery method that combines changepoint detection and structural vector auto-regressive model. It uses the changepoint detection method to identify the time point of change, then divides the time of the previous step into stationary intervals, and further uses the stationary algorithm to infer their local causal structures. Experiments on simulated and real-world data prove the effectiveness of the proposed method.

Key words: causal discovery, non-stationary, causal network

郝志峰, 张维杰, 蔡瑞初, 陈薇. 基于条件独立性检验的非稳态因果发现方法[J]. 计算机工程与应用, 2024, 60(10): 113-120.

HAO Zhifeng, ZHANG Weijie, CAI Ruichu, CHEN Wei. Non-Stationary Causal Discovery Method Based on Conditional Independence Test[J]. Computer Engineering and Applications, 2024, 60(10): 113-120.

参考文献

[1] RUNGE J, BATHIANY S, BOLLT E, et al. Inferring causation from time series in earth system sciences[J]. Nature Communications, 2019, 10(1): 1-13.
[2] 蔡瑞初, 陈薇, 张坤, 等. 基于非时序观察数据的因果关系发现综述[J]. 计算机学报, 2017, 40(6): 1470-1490.
CAI R C, CHEN W, ZHANG K, et al. A survey on non-temporal series observational data based causal discovery[J]. Chinese Journal of Computers, 2017, 40(6): 1470-1490.
[3] RUNGE J, NOWACK P, KRETSCHMER M, et al. Detecting and quantifying causal associations in large nonlinear time series datasets[J]. Science Advances, 2019, 5(11): 4996.
[4] SHIMIZU S, HOYER P O, HYV?RINEN A, et al. A linear non-Gaussian acyclic model for causal discovery[J]. Journal of Machine Learning Research, 2006, 7(10): 2003-2030.
[5] HYV?RINEN A, SHIMIZU S, HOYER P O. Causal modelling combining instantaneous and lagged effects: an identifiable model based on non-Gaussianity[C]//Proceedings of the 25th International Conference on Machine Learning, 2008: 424-431.
[6] HOYER P, JANZING D, MOOIJ J M, et al. Nonlinear causal discovery with additive noise models[C]//Advances in Neural Information Processing Systems 21, 2008: 689-696.
[7] PETERS J, JANZING D, SCH?LKOPF B. Causal inference on time series using restricted structural equation models[C]//Advances in Neural Information Processing Systems 26, 2013: 154-162.
[8] CHU T, GLYMOUR C, RIDGEWAY G. Search for additive nonlinear time series causal models[J]. Journal of Machine Learning Research, 2008, 9(5): 967-991.
[9] HUANG B, ZHANG K, ZHANG J, et al. Causal discovery from heterogeneous/nonstationary data[J]. The Journal of Machine Learning Research, 2020, 21(1): 3482-3534.
[10] HUANG B, ZHANG K, GONG M, et al. Causal discovery and forecasting in nonstationary environments with state-space models[C]//Proceedings of the 36th International Conference on Machine Learning, 2019: 2901-2910.
[11] KILLICK R, FEARNHEAD P, ECKLEY I A. Optimal detection of changepoints with a linear computational cost[J]. Journal of the American Statistical Association, 2012, 107(500): 1590-1598.
[12] SPIRTES P, GLYMOUR C N, SCHEINES R, et al. Causation, prediction, and search[M]. Cambridge: The MIT Press, 2001.
[13] VERMA T S, PEARL J. Equivalence and synthesis of causal models[C]//Proceedings of the 6th Annual Conference on Uncertainty in Artificial Intelligence, 1990: 255-270.
[14] JACKSON B, SCARGLE J D, BARNES D, et al. An algorithm for optimal partitioning of data on an interval[J]. IEEE Signal Processing Letters, 2005, 12(2): 105-108.
[15] R?SLER O, SUENDERMANN D. A first step towards eye state prediction using eeg[C]//Proceedings of the 2013 International Conference on Applied Informatics for Health and Life Sciences, Istanbul, 2013: 3-4.
[16] NOLTE G, ZIEHE A, NIKULIN V V, et al. Robustly estimating the flow direction of information in complex physical systems[J]. Physical Review Letters, 2008, 100(23): 234101.

基于条件独立性检验的非稳态因果发现方法

Non-Stationary Causal Discovery Method Based on Conditional Independence Test

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

[1]	董元发1，2，晋良海1，燕乔1. 混凝土坝非稳态施工过程异常监控与管理研究[J]. 计算机工程与应用, 2017, 53(20): 208-213.
[2]	朱亚光，马存宝，朱道德，周伟. 概率因果网络在航电系统多故障诊断中的应用[J]. 计算机工程与应用, 2010, 46(7): 242-244.
[3]	马存宝¹,裴林¹,李文娟¹,史浩山². 改进的关联模式竞争求解算法[J]. 计算机工程与应用, 2009, 45(4): 205-207.