Causal Discovery from Conditionally Stationary Time Series

2021/10/12 by Carles Balsells Rodas, Carles Balsells-Rodas, Balsells-Rodas, Carles +12 · 1 voice
Computer Science · Mathematics · #Bayesian Modeling and Causal Inference #Data Stream Mining Techniques #Explainable Artificial Intelligence (XAI) #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.2110.06257

arxiv published 2021/10/12 · arxiv updated 2025/06/05

Abstract

Causal discovery, i.e., inferring underlying causal relationships from observational data, is highly challenging for AI systems. In a time series modeling context, traditional causal discovery methods mainly consider constrained scenarios with fully observed variables and/or data from stationary time-series. We develop a causal discovery approach to handle a wide class of nonstationary time series that are conditionally stationary, where the nonstationary behaviour is modeled as stationarity conditioned on a set of latent state variables. Named State-Dependent Causal Inference (SDCI), our approach is able to recover the underlying causal dependencies, with provable identifiablity for the state-dependent causal structures. Empirical experiments on nonlinear particle interaction data and gene regulatory networks demonstrate SDCI's superior performance over baseline causal discovery methods. Improved results over non-causal RNNs on modeling NBA player movements demonstrate the potential of our method and motivate the use of causality-driven methods for forecasting.

Citations

• Auto-Encoding Variational Bayes
• Macroeconomics and Reality
• Investigating Causal Relations by Econometric Models and Cross-spectral Methods
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• DAGs with NO TEARS: Continuous Optimization for Structure Learning
• Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data

Discussions

• I was at the Symposium on Causality in Biology & AI in Barcelona. I presented a flash talk on inferring time-varying gene regulatory networks from time series scRNA-Seq data. Some results here - arxiv [bsky, 3 points, 1 comments]

Related

• Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models
• Causal Discovery in Multivariate Time Series through Mutual Information Featurization
• Identifying Nonstationary Causal Structures with High-Order Markov Switching Models
• Causal Structure Learning for Dynamical Systems with Theoretical Score Analysis
• Causal Discovery on Irregular Time Series