arXiv Machine Learning
State-Space Unlearning for Non-Stationary Bias in Land Surface Forecasting
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
3d ago
Desertification Forecasting in Semi-Arid Regions: Using Deep Learning to Probe the Observability Limits of Land Degradation Dynamics
The study trains a convolutional LSTM on ten annual Earth Observation variables to forecast desertification dynamics in Ceará, Brazil, achieving modest improvements over persistence and capturing 53‑69% of the regional degradation trend. While the model slightly outperforms simple baselines one to four years ahead, it explains only a small fraction of cell‑level variance, indicating that EO data are more reliable for regional trends than for individual pixel predictions. The authors also demonstrate that elevation and EVI channels degrade forecast skill, and use Monte Carlo dropout to quantify uncertainty in future degradation maps.
arXiv Machine Learning
Jul 8
Regularity and Stability Properties of Selective SSMs with Discontinuous Gating
arXiv:2505. 11602v3 Announce Type: replace Abstract: Selective State-Space Models (SSMs) such as Mamba have become central to long-sequence modeling.
arXiv AI
Sep 16
Schema-Adaptive Action-Conditioned JEPA for Cross-Machine CNC Transfer under Partial Sensor Overlap
The paper introduces a schema‑adaptive action‑conditioned Joint‑Embedding Predictive Architecture (SAAC‑JEPA) for cross‑machine CNC transfer when only a subset of sensors overlap between source and target machines. Experiments show that pretraining does not improve source‑only forecasting, but a carefully selected action‑conditioned JEPA model achieves a zero‑shot RMSE of 0.546 on the target, outperforming persistence but falling short of certain baseline models. Ablation studies reveal that adding RevIN improves RMSE but harms calibration, and limited post‑lock adaptation can further reduce error.
arXiv Machine Learning
Jul 10
Scalable and Trustworthy Earth Observation Foundation Models
arXiv:2607. 07758v1 Announce Type: new Abstract: Foundation models (FMs) have transformed machine learning from isolated task-specific model development toward general-purpose models pretrained on broad data and adapted to multiple downstream tasks.
arXiv Machine Learning
Jun 26
Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather Covariates
arXiv:2602. 17683v3 Announce Type: replace Abstract: Short-term forecasting of vegetation dynamics is a key enabler for data-driven decision support in precision agriculture.
arXiv Machine Learning
Aug 24
Free-Probability Kernels for Zero-Rollout Hyperparameter Selection in Reservoir Computing
arXiv:2608.20998v1 Announce Type: new Abstract: Reservoir computing (RC) couples a fixed recurrent dynamical system with a trained lightweight readout, but this efficiency is partly lost during hyper...