Research map / Probabilistic and scientific modeling / Causal discovery
SCDM: Scalable Causal Discovery in Nonlinear Temporal Systems with Meta-Learning
Jingbo Wang, Kegeng Tang, Zihao Wang, Shaogang Ren
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes SCDM, a shared-parameter meta-learning framework for scalable causal discovery in high-dimensional nonlinear multivariate time series.
Problem
Causal discovery from nonlinear multivariate time series in high-dimensional systems, where candidate directed relations grow quadratically with the number of variables; existing methods suffer from target-wise fitting, repeated conditional testing, or expensive graph search.
Approach
A shared-parameter meta-learning framework that performs temporal causal discovery in a scalable way, avoiding per-target model fitting, repeated conditional tests, and graph-search procedures.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Subsample target tasks B=32 instead of all targets (B=100) (ablation) | SCDM · Simulated fMRI 100 variables | training time (s): 2049.9 -> 668.4 | Improved |
| Subsample B=32 instead of B=100 (ablation) | SCDM · Simulated fMRI 100 variables | AUPRC: 0.552 -> 0.728 | Improved |
| Reduce task batch from 32 to 10 (ablation) | SCDM · Simulated fMRI 100 variables | AUPRC: 0.728 -> 0.656 | Hurt |
| Use support/query meta-objective instead of pooled multi-task training (ablation) | SCDM · Held-out target adaptation | query MSE: 4.305 -> 3.782 | Improved |
| Add inner adaptation steps (K>0) versus K=0 (ablation) | SCDM · Held-out target adaptation | query MSE: 5.487 -> 3.782 | Improved |
| Add shared causal-strength matrix W (ablation) | SCDM · Simulated fMRI | AUPRC: 0.390 -> 0.728 | Improved |
| Use structural readout instead of source-isolated readout (ablation) | SCDM · Simulated fMRI | AUPRC: 0.693 -> 0.728 | Improved |
| Use source-isolated readout instead of structural readout (ablation) | SCDM · Simulated EEG | AUPRC: 0.426 -> 0.476 | Improved |
| Use SCDM instead of UnCLe on NC8 | SCDM · NC8 | AUROC: 0.947 -> 0.916 | Hurt |
| Use SCDM instead of GVAR on simulated EEG | SCDM · Simulated EEG | AUROC: 0.750 -> 0.717 | Hurt |
| Use SCDM instead of UnCLe on ROI simulated fMRI | SCDM · Simulated fMRI (ROI-level) | AUROC: 0.661 -> 0.867 | Improved |
| Use SCDM on 1000-variable fMRI at 200 s budget | SCDM · Voxel-level fMRI (d=1000), 200 s | AUPRC enrichment: 2.05x -> 1.96x | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- UnCLe — UnCLe: Towards Scalable Dynamic Causal Discovery in Non-linear Temporal Systems (NeurIPS 2025) · better on 4 of 6 shared results
- CUTS+ — CUTS: Neural Causal Discovery from Irregular Time-Series Data (ICLR 2023) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- UnCLe: Towards Scalable Dynamic Causal Discovery in Non-linear Temporal Systems (NeurIPS 2025 · similarity 0.90)
- A Meta-Learning Approach to Bayesian Causal Discovery (ICLR 2025 · similarity 0.89)
- Differentiable Causal Discovery for Latent Hierarchical Causal Models (ICLR 2025 · similarity 0.89)
- CausalTime: Realistically Generated Time-series for Benchmarking of Causal Discovery (ICLR 2024 · similarity 0.89)
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.89)
- CUTS+: High-Dimensional Causal Discovery from Irregular Time-Series (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
