TTCD:Transformer Integrated Temporal Causal Discovery from Non-Stationary Time Series Data

TL;DR

The paper introduces TTCD, a transformer-based framework for causal discovery in non-stationary time series, effectively handling complex, noisy, and evolving data without strong assumptions.

Contribution

It presents a novel end-to-end method combining a non-stationary feature learner and causal structure learner with reconstruction-guided signal distillation.

Findings

TTCD outperforms state-of-the-art methods in synthetic and real datasets.

The approach effectively handles non-stationarity and noise.

Experimental results demonstrate high accuracy and consistency with domain knowledge.

Abstract

The widespread availability of complex time series data in various domains such as environmental science, epidemiology, and economics demands robust causal discovery methods that can identify intricate contemporaneous and lagged relationships in non-stationary, nonlinear, and noisy settings. Existing constraint-based methods often rely heavily on conditional independence tests that degrade for limited data samples and complex distributions, while score-based methods impose strong statistical assumptions. Recent methods address special cases such as change point detection or distribution shifts, but struggle to provide a unified solution. We propose the Transformer Integrated Temporal Causal Discovery (TTCD) Framework, a novel end-to-end approach that learns contemporaneous and lagged causal relations from non-stationary time series. TTCD introduces a Non-Stationary Feature Learner integrating temporal and frequency-domain attention with dynamic non-stationarity profiling, and a custom Causal Structure Learner. A key innovation is reconstruction-guided causal signal distillation, to distill essential causal signals through the reconstruction process of the transformer decoder, which mitigates noise and spurious correlations while preserving meaningful dependencies. The Causal Structure Learner operates on distilled reconstructed signals to infer the underlying causal graph without restrictive assumptions on noise distributions or data generation processes. Experiments on synthetic, benchmark, and real world datasets show that TTCD consistently outperforms state-of-the-art baselines in both accuracy and consistency with domain knowledge, demonstrating the approach's effectiveness for causal discovery in challenging real world contexts.