Research map / Probabilistic and scientific modeling / Causal discovery
Neural Granger Causality
Alex Tank, Ian Covert, Nicholas J. Foti, Ali Shojaie, Emily B. Fox
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use hierarchical penalty instead of group lasso penalty (ablation) | cMLP · Lorenz-96 (F=20, T=750), K=5 | AUROC (%): 88.1 -> 95.5 | Improved |
| use hierarchical penalty instead of group lasso penalty (ablation) | cMLP · Lorenz-96 (F=20, T=750), K=20 | AUROC (%): 80.5 -> 95.2 | Improved |
| use hierarchical penalty instead of group sparse group lasso penalty (ablation) | cMLP · Lorenz-96 (F=20, T=750), K=20 | AUROC (%): 83.3 -> 95.2 | Improved |
| increase maximum lag K from 5 to 20 with group penalty | cMLP GROUP · Lorenz-96 (F=20, T=750) | AUROC (%): 88.1 -> 80.5 | Hurt |
| increase maximum lag K from 5 to 20 with hierarchical penalty | cMLP HIER · Lorenz-96 (F=20, T=750) | AUROC (%): 95.5 -> 95.2 | No clear effect |
| increase hidden units H from 5 to 100 (ablation) | cMLP · Lorenz-96 (F=10, T=1000) | AUROC (%): 96.5 -> 98.5 | Improved |
| increase hidden units H from 5 to 100 (ablation) | cMLP · Lorenz-96 (F=40, T=1000) | AUROC (%): 91.0 -> 94.5 | Improved |
| increase hidden units H from 5 to 100 (ablation) | cLSTM · Lorenz-96 (F=10, T=1000) | AUROC (%): 91.9 -> 95.7 | Improved |
| increase hidden units H from 5 to 100 (ablation) | cLSTM · Lorenz-96 (F=40, T=1000) | AUROC (%): 86.9 -> 95.2 | Improved |
| use Adam instead of ISTA (ablation) | cMLP · Lorenz-96 (F=40, T=1000) | AUROC (%): 94.1 -> 95.1 | No clear effect |
| use GISTA line search instead of ISTA (ablation) | cMLP · Lorenz-96 (F=10, T=1000) | AUROC (%): 98.0 -> 98.0 | No clear effect |
| increase series length T from 250 to 1000 | cMLP · Lorenz-96 (F=10) | AUROC (%): 86.6 -> 98.4 | Improved |
Closest papers by meaning
- Jacobian Regularizer-based Neural Granger Causality (ICML 2024 · similarity 0.90)
- Exploring Neural Granger Causality with xLSTMs: Unveiling Temporal Dependencies in Complex Data (NeurIPS 2025 · similarity 0.89)
- TNPAR: Topological Neural Poisson Auto-Regressive Model for Learning Granger Causal Structure from Event Sequences (AAAI 2024 · similarity 0.89)
- Towards the Disappearing Truth: Fine-Grained Joint Causal Influences Learning with Hidden Variable-Driven Causal Hypergraphs in Time Series (AAAI 2024 · similarity 0.87)
- CUTS+: High-Dimensional Causal Discovery from Irregular Time-Series (AAAI 2024 · similarity 0.87)
- GCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger Causality (AAAI 2025 · similarity 0.86)
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.
