Intervention-Based Time Series Causal Discovery via Simulator-Generated Interventional Distributions
About
We propose SVAR-FM (Structural VAR with Flow Matching), a framework for time series causal discovery that treats a physics-based simulator as a mechanical realization of Pearl's do operator. Clamping a variable inside the simulator physically severs confounding paths, producing interventional data by construction. Conditional Flow Matching then learns the nonlinear interventional conditionals. Theoretically, we prove that the full structural VAR becomes identifiable under a coverage condition on the simulator-clampable variables, and derive an end-to-end error bound that decomposes into Monte Carlo, simulator fidelity, and Flow Matching terms. A sign-flip corollary predicts that when simulator accuracy falls below a threshold, the estimated causal effect reverses sign. Empirically, a benchmark across four scientific domains confirms that SVAR-FM recovers the correct causal sign where observational methods produce sign-reversed estimates due to confounding. A case study in ultrafast laser physics verifies the sign-flip prediction by physically varying the accuracy level of a first-principles quantum solver: the low-accuracy setting reverses the causal sign, while the high-accuracy setting recovers the correct direction (R-squared = 0.983, zero bias).
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Causal Discovery | CausalDynamics Lorenz | AUROC86 | 9 | |
| Causal Discovery | CausalDynamics Rössler | AUROC80 | 9 | |
| Causal Discovery | Tigramite S2 Non-Linear Gaussian | F1 Score84 | 8 | |
| Causal Discovery | Tigramite S4 Non-Linear Non-Gaussian | F1 Score92 | 8 | |
| Causal Discovery | Tigramite S6 Latent Variables | F1 Score100 | 8 | |
| Causal Discovery | CausalDynamics CoupledLorenz | AUROC76.4 | 8 | |
| Causal effect estimation | CausalSim-Diabetes | Sign Correct100 | 8 | |
| Causal Discovery | Tigramite S1 Linear Gaussian | F1 Score100 | 8 | |
| Causal Discovery | Tigramite S3 Linear Non-Gaussian | F1 Score100 | 8 | |
| Causal Discovery | Tigramite S5 Contemporaneous | F1 Score79.3 | 8 |