用规则解释气候模型，让极端天气预测更透明。
SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events
- 基于地理位置调制输入，提升气候数据的语义捕捉能力
- 在火灾、台风和大气河预测中实现高精度特征解释
- 生成符合科学文献的可读规则，适合气候研究者使用
大规模气象与气候（W&C）数据为建模极端地球事件（ExEE）及其影响提供了新机遇，但其在实际应用中的推广受限于模型缺乏可解释性。尽管在文本和图像领域，稀疏自编码器（SAEs）已证明能有效提取人类可理解的概念，但因其特性，该方法在气候数据上的应用仍具挑战。为此，我们提出：(i) 基于地理坐标的输入调制机制，以捕捉环境模式的局部语义；(ii) 构建一组基于规则的 SAE-Xplainer，用于解释复杂多模态环境预测因子产生的高维特征。我们在三类极端事件——火灾预测、热带气旋检测和大气河流识别上评估该方法。结果表明，输入调制显著提升了重建性能与特征利用率；SAE-Xplainers 能将复杂气候模式分解为与科学文献一致的人类可理解规则，并支持特征吸收的识别。
原文摘要 · Abstract (English)
The emergence of large-scale Weather and Climate (W&C) datasets offers new opportunities for modeling extreme Earth events (ExEE) and their impacts using deep learning. However, their adoption in operational settings remains limited by the lack of models' interpretability. While for conventional text and image modalities, tools such as Sparse Autoencoders (SAEs) have proven effective for extracting human-understandable concepts, their use for the analysis of ExEE remains challenging due to the nature of W&C data. To address this, we introduce (i) a geographic location-based modulation of the inputs of SAE to capture the local semantic meaning of environmental patterns, and (ii) an ensemble of rule-based SAE-Xplainers to interpret the resulting high-dimensional features derived from complex, multi-modal environmental predictors. We evaluate our method on three ExEE types: the prediction of fires, and the detection of tropical cyclones and atmospheric rivers. We show that SAE input modulation improves both reconstruction performance and feature utilization, and that our SAE-Xplainers enable faithful interpretation of complex climatic patterns by unfolding them into human-understandable rules that are consistent with the scientific literature, while also supporting the identification of feature absorption.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。