Research map / Time series / Causal and generative time series
COGS: A Causal Representation Learning Framework for Out-of-Distribution Generalization in Time Series
Xinxin Song, Yuxiao Cheng, Tingxiong Xiao, Jinli Suo
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 |
|---|---|---|---|
| add temporal causal representation learning loss (latent identification with prior network) (ablation) | COGS (ERM backbone) · PAMAP | average accuracy: 48.62 -> 53.75 | Improved |
| add temporal causal representation learning loss (ablation) | COGS · WESAD | average accuracy: 47.41 -> 50.37 | Improved |
| add causal graph discovery and causally invariant prediction loss (ablation) | COGS · PAMAP | average accuracy: 53.75 -> 55.88 | Improved |
| add causal graph discovery and causally invariant prediction loss (ablation) | COGS · USC-HAD | average accuracy: 66.69 -> 67.41 | Improved |
| add prototype-guided domain inference (L_EI) replacing predefined domain labels (ablation) | COGS · PAMAP | average accuracy: 55.88 -> 60.08 | Improved |
| add prototype-guided domain inference (L_EI) (ablation) | COGS · WESAD | average accuracy: 51.96 -> 58.09 | Improved |
| add prototype-guided domain inference (L_EI) (ablation) | COGS · USC-HAD | average accuracy: 67.41 -> 71.18 | Improved |
| COGS vs ERM | COGS · PAMAP | average accuracy: 48.62 -> 60.68 | Improved |
| COGS vs best baseline | COGS · DSADS | average accuracy: 82.55 -> 83.91 | Improved |
| COGS vs best baseline | COGS · WESAD | average accuracy: 56.58 -> 59.04 | Improved |
Closest papers by meaning
- Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains (ICML 2025 · similarity 0.87)
- On the Identification of Temporal Causal Representation with Instantaneous Dependence (ICLR 2025 · similarity 0.87)
- Structured Temporal Causality for Interpretable Multivariate Time Series Anomaly Detection (NeurIPS 2025 · similarity 0.87)
- CauDiTS: Causal Disentangled Domain Adaptation of Multivariate Time Series (ICML 2024 · similarity 0.86)
- Discovering Mixtures of Structural Causal Models from Time Series Data (ICML 2024 · similarity 0.86)
- Time-Series Forecasting for Out-of-Distribution Generalization Using Invariant Learning (ICML 2024 · 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.
