CaDrift: A Time-dependent Causal Generator of Drifting Data Streams
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914346453434368 |
|---|---|
| author | Barboza, Eduardo V. L. Barddal, Jean Paul Sabourin, Robert Cruz, Rafael M. O. |
| author_facet | Barboza, Eduardo V. L. Barddal, Jean Paul Sabourin, Robert Cruz, Rafael M. O. |
| contents | This work presents Causal Drift Generator (CaDrift), a time-dependent synthetic data generator framework based on Structural Causal Models (SCMs). The framework produces a virtually infinite combination of data streams with controlled shift events and time-dependent data, making it a tool to evaluate methods under evolving data. CaDrift synthesizes various distributional and covariate shifts by drifting mapping functions of the SCM, which change underlying cause-and-effect relationships between features and the target. In addition, CaDrift models occasional perturbations by leveraging interventions in causal modeling. Experimental results show that, after distributional shift events, the accuracy of classifiers tends to drop, followed by a gradual retrieval, confirming the generator's effectiveness in simulating shifts. The framework has been made available on GitHub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20329 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CaDrift: A Time-dependent Causal Generator of Drifting Data Streams Barboza, Eduardo V. L. Barddal, Jean Paul Sabourin, Robert Cruz, Rafael M. O. Machine Learning Databases This work presents Causal Drift Generator (CaDrift), a time-dependent synthetic data generator framework based on Structural Causal Models (SCMs). The framework produces a virtually infinite combination of data streams with controlled shift events and time-dependent data, making it a tool to evaluate methods under evolving data. CaDrift synthesizes various distributional and covariate shifts by drifting mapping functions of the SCM, which change underlying cause-and-effect relationships between features and the target. In addition, CaDrift models occasional perturbations by leveraging interventions in causal modeling. Experimental results show that, after distributional shift events, the accuracy of classifiers tends to drop, followed by a gradual retrieval, confirming the generator's effectiveness in simulating shifts. The framework has been made available on GitHub. |
| title | CaDrift: A Time-dependent Causal Generator of Drifting Data Streams |
| topic | Machine Learning Databases |
| url | https://arxiv.org/abs/2602.20329 |