Metraipolis › archive › dc_dcdd3f6352ee

Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction — arXiv:2609.17763

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.17763
License (as archived): CC0-1.0
Archived: 2026-10-15T15:04:25.486Z
SHA-256 of the archived text: b008e5020d28edb409da50abaa97439c201a050537412c81675b1c649c29b1c9
Stable URL of this page: https://metraipolis.com/documento/dc_dcdd3f6352ee

How to cite

Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction — arXiv:2609.17763 — arXiv (en), https://arxiv.org/abs/2609.17763 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dcdd3f6352ee (Metraipolis archive dc_dcdd3f6352ee, sha256 b008e5020d28edb409da50abaa97439c201a050537412c81675b1c649c29b1c9)

BibTeX

@misc{dcdcdd3f6352ee,
  title = {Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction — arXiv:2609.17763},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17763},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dcdd3f6352ee; sha256 b008e5020d28edb409da50abaa97439c201a050537412c81675b1c649c29b1c9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17763. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Next-day wildfire prediction requires models whose forecasts can be evaluated alongside the assumptions and historical evidence used in their computation. Although deep learning can learn spatial patterns from remote-sensing data, predictive performance alone does not establish physical fidelity or operational trustworthiness. This study investigates three modular augmentations for next-day active-fire prediction: wind- and slope-conditioned attention biases, physics-feature retrieval-augmented output correction, and fire conditioned dual-stream gating. The attention biases expose prescribed directional preferences, while the retrieval module selects historical tiles using a nine-dimensional environmental and fire-state descriptor and applies a learned correction to a frozen model's logits. The modules are evaluated across five backbones on the Next Day Wildfire Spread benchmark, using staged ablations, directional audits, retrieval perturbations, calibration measures, and computational comparisons. The three-seed mean F1 score and area under the precision--recall curve (AUC-PR) of a SwinUNETR model with all three augmentations are 0.4216 and 0.3673. Then, a mixed ensemble (two augmented architectures and one non-augmented architecture) model achieves 0.4292 and 0.3790. Benefits vary across architectures, and retrieval-related improvements in AUC-PR do not consistently translate into higher F1. The constructed wind bias aligns closely with input wind, but its alignment with observed next-day fire displacement is much weaker, distinguishing prior inspectability from predictive physical fidelity. The study contributes a framework for exposing and evaluating selected domain-informed components within wildfire prediction models. Together, the results presented show that predictive performance, operational trustworthiness, and computational practicality need not be competing objectives.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.17763 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T14:08:19.066Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).