Causal Discovery from Conditionally Stationary Time Series

TY - JOUR

T1 - Causal Discovery from Conditionally Stationary Time Series

AU - Balsells-Rodas, Carles

AU - Sumba, Xavier

AU - Narendra, Tanmayee

AU - Tu, Ruibo

AU - Schweikert, Gabriele

AU - Kjellström, Hedvig

AU - Li, Yingzhen

N1 - Publisher Copyright: © 2025, ML Research Press. All rights reserved.

PY - 2025

Y1 - 2025

N2 - 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.

AB - 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.

UR - https://www.scopus.com/pages/publications/105023641342

M3 - Conference article

AN - SCOPUS:105023641342

VL - 267

SP - 2715

EP - 2741

JO - Proceedings of Machine Learning Research

JF - Proceedings of Machine Learning Research

T2 - 42nd International Conference on Machine Learning, ICML 2025

Y2 - 13 July 2025 through 19 July 2025

ER -