Novel Kolmogorov-Arnold network architectures for accurate flood susceptibility mapping: a comparative study
Bibliographic record
Abstract
Accurate mapping of flood susceptibility (FSM) is of paramount importance for the effective management and mitigation of this deadly disaster. This study introduces a novel framework based on the Kolmogorov-Arnold Network (KAN) for enhanced FSM, which was applied to two basins in Iran: the Karun and Gorganrud basins. Three KAN-based models were implemented and evaluated. The performance of the Boubaker-KAN, Cheby-KAN, and VietaPell-KAN models was evaluated in comparison to state-of-the-art machine learning techniques, including Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Extreme Gradient Boosting (XGBoost), and Categorical Boosting (CatBoost). All models were trained using a set of conditioning factors related to flooding, including topographical indicators, land cover data, and soil characteristics. The delineation of flood-prone areas was conducted through the identification of historical inundation incidents, as observed in satellite imagery and documented in official reports. The results demonstrate that KAN-based models exhibit superior performance, with an average overall accuracy of 92.5 % across the two basins. Furthermore, the KAN models achieved an average F1 score of 93.90 % and an average Matthews Correlation Coefficient (MCC) of 0.866, demonstrating superior performance in these key metrics in comparison to other techniques. The superior performance of KAN-based models can be attributed to their capacity to capture intricate, non-linear relationships between flood conditioning factors and flood occurrences, as per the Kolmogorov-Arnold representation theorem. A visual comparison of the flood susceptibility maps demonstrates that KAN models effectively capture the subtle topographical and hydrological features that contribute to localized flooding. This research contributes to the advancement of FSM techniques, offering improved tools for flood risk assessment and management. Future work should focus on incorporating additional dynamic variables and exploring hybrid approaches combining KAN architectures with ensemble methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".