Research map / Probabilistic and scientific modeling / Causal discovery
Learning General Causal Structures with Hidden Dynamic Process for Climate Analysis
Minghao Fu, Biwei Huang, Zijian Li, Yujia Zheng, Ignavier Ng, Guangyi Chen, Yingyao Hu, Kun Zhang
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 |
|---|---|---|---|
| model observed-to-observed causal edges and noisy generation (CaDRe) versus temporal CRL baselines | synthetic climate data · Independent latent setting | MCC: 0.9106 -> 0.9811 | Improved |
| CaDRe versus temporal CRL baselines under sparse latent dynamics | synthetic climate data · Sparse latent setting | MCC: 0.7701 -> 0.9306 | Improved |
| CaDRe versus temporal CRL baselines under dense latent dynamics | synthetic climate data · Dense latent setting | MCC: 0.6714 -> 0.6750 | Improved |
| CaDRe versus constraint-based CD (PC, FCI, CD-NOD, PCMCI, LPCMCI) | synthetic data, varying sample size and dimension · Synthetic causal discovery | SHD, precision, recall, F1 | Improved |
| CaDRe versus neural/constraint CD baselines on wind-field surrogate | CESM2 · CESM2 causal discovery | WSHD: 0.019 -> 0.012 | Improved |
| CaDRe one-step graph generation versus constraint-based methods | CESM2 · CESM2 causal discovery | Latency (ms): 3508.50 -> 1.095 | Improved |
| CaDRe versus forecasting baselines at horizon 96 | CESM2 · CESM2 forecasting | MSE: 0.409 -> 0.410 | No clear effect |
| CaDRe versus forecasting baselines at long horizon | Weather · Weather forecasting 336 | MSE: 0.316 -> 0.270 | Improved |
| CaDRe versus Timer-XL at short horizon | ERSST · ERSST forecasting 96 | MSE: 0.163 -> 0.145 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- CARD — CARD: Classification and Regression Diffusion Models (NeurIPS 2022) · better on 1 of 2 shared results
- CaRiNG — CaRiNG: Learning Temporal Causal Representation under Non-Invertible Generation Process (ICML 2024) · better on 1 of 1 shared result
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
- SPACETIME: Causal Discovery from Non-Stationary Time Series (AAAI 2025 · similarity 0.89)
- Marrying Causal Representation Learning with Dynamical Systems for Science (NeurIPS 2024 · similarity 0.89)
- Causal Climate Emulation with Bayesian Filtering (NeurIPS 2025 · similarity 0.89)
- Discovering Latent Causal Graphs from Spatiotemporal Data (ICML 2025 · similarity 0.88)
- CausalDynamics: A large‐scale benchmark for structural discovery of dynamical causal models (NeurIPS 2025 · similarity 0.88)
- Causal discovery with endogenous context variables (NeurIPS 2024 · similarity 0.87)
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.
