SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events

• URL: http://arxiv.org/abs/2608.20117v1
• Date: Thu, 20 Aug 2026 14:49:26 GMT
• Title: SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events
• Abstract summary: Large-scale Weather and Climate (W&C) datasets offer new opportunities for modeling extreme Earth events (ExEE)<n>Their adoption in operational settings remains limited by the lack of models' interpretability.<n>We introduce (i) a geographic location-based modulation of the inputs of Sparse Autoencoders (SAEs) 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.
• Score: 27.512209212824757
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: 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.

Related papers

• Probing Geospatial SSL Representations with Environmental Signals [51.908606610432194]
  Self-supervised learning (SSL) is designed to learn generic, transferable representations rather than representations optimized for a single task.<n>We ask a different question: do SSL representations preserve statistical associations with environmental variables that co-vary with the imaging process?<n>To answer this question, we probe SSL representations using co-located ERA5 reanalysis variables, a global dataset of physically consistent environmental variables.
  arXiv  Detail & Related papers  (2026-07-06T15:24:25Z)
• The Rate-Distortion-Polysemanticity Tradeoff in SAEs [24.387030280550732]
  This paper introduces the Rate-Distortion-Polysemanticity tradeoff in SAEs.<n>We show that restricting the SAE to be monosemantic necessarily comes with an increase in rate and distortion.<n>We extend the analysis to real-world settings by deriving necessary conditions that a polysemanticity measure should satisfy.
  arXiv  Detail & Related papers  (2026-05-14T11:10:52Z)
• Earth-o1: A Grid-free Observation-native Atmospheric World Model [81.55899748753434]
  We present Earth-o1, an observation-native atmospheric world model that overcomes structural limitations.<n>By integrating diverse sensor inputs into a unified, grid-free dynamical field, the model autonomously advances the atmospheric state in space and time.<n>We show that this fundamentally distinct paradigm enables direct, real-time forecasting and cross-sensor inference without the overhead of explicit numerical solvers.
  arXiv  Detail & Related papers  (2026-05-07T14:27:48Z)
• Toward generative machine learning for boosting ensembles of climate simulations [0.0]
  We develop a conditional Variational Autoencoder (cVAE) trained on a limited sample of climate simulations to generate arbitrary large ensembles.<n>We show that the cVAE model learns the underlying distribution of the data and generates physically consistent samples that reproduce realistic low and high moment statistics.
  arXiv  Detail & Related papers  (2026-02-06T00:54:19Z)
• DAWP: A framework for global observation forecasting via Data Assimilation and Weather Prediction in satellite observation space [60.729377189859]
  We propose our DAWP framework to enable AIWPs to operate in a complete observation space.<n>AIDA module applies a mask multi-modality autoencoder for assimilating irregular satellite observation tokens.<n>We show that AIDA significantly improves the roll out and efficiency of AIWP and holds promising potential to be applied in global precipitationresolution forecasting.
  arXiv  Detail & Related papers  (2025-10-13T03:13:35Z)
• Knowledge-Guided Adaptive Mixture of Experts for Precipitation Prediction [9.508961267296579]
  We propose an Adaptive Mixture of Experts (MoE) model tailored for precipitation rate prediction.<n>Each expert within the model specializes in a specific modality or precipitation-temporal pattern.<n>We introduce an interactive web-based visualization tool that enables users to intuitively explore historical weather patterns over time and space.
  arXiv  Detail & Related papers  (2025-09-14T22:31:46Z)
• Open-set Anomaly Segmentation in Complex Scenarios [88.11076112792992]
  This paper introduces ComsAmy, a benchmark for open-set anomaly segmentation in complex scenarios.<n>ComsAmy encompasses a wide spectrum of adverse weather conditions, dynamic driving environments, and diverse anomaly types.<n>We propose a novel energy-entropy learning (EEL) strategy that integrates the complementary information from energy and entropy.
  arXiv  Detail & Related papers  (2025-04-28T12:00:10Z)
• Leveraging Land Cover Priors for Isoprene Emission Super-Resolution [15.868193361155656]
  This study contributes to atmospheric chemistry and climate modeling by providing a cost-effective, data-driven approach to refining BVOC emission maps.<n>The proposed method enhances the usability of satellite-based emissions data, supporting applications in air quality forecasting, climate impact assessments, and environmental studies.
  arXiv  Detail & Related papers  (2025-03-24T13:23:46Z)
• Towards Causal Representations of Climate Model Data [18.82507552857727]
  This work delves into the potential of causal representation learning, specifically the emphCausal Discovery with Single-parent Decoding (CDSD) method. Our findings shed light on the challenges, limitations, and promise of using CDSD as a stepping stone towards more interpretable and robust climate model emulation.
  arXiv  Detail & Related papers  (2023-12-05T16:13:34Z)
• FREE: The Foundational Semantic Recognition for Modeling Environmental Ecosystems [56.0640340392818]
  We introduce a framework, FREE, that enables the use of varying features and available information to train a universal model.<n>The core idea is to map available environmental data into a text space and then convert the traditional predictive modeling task in environmental science to a semantic recognition problem.<n>Our evaluation on two societally important real-world applications, stream water temperature prediction and crop yield prediction, demonstrates the superiority of FREE over multiple baselines.
  arXiv  Detail & Related papers  (2023-11-17T00:53:09Z)
• Conditioned Human Trajectory Prediction using Iterative Attention Blocks [70.36888514074022]
  We present a simple yet effective pedestrian trajectory prediction model aimed at pedestrians positions prediction in urban-like environments. Our model is a neural-based architecture that can run several layers of attention blocks and transformers in an iterative sequential fashion. We show that without explicit introduction of social masks, dynamical models, social pooling layers, or complicated graph-like structures, it is possible to produce on par results with SoTA models.
  arXiv  Detail & Related papers  (2022-06-29T07:49:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.