mlsanityData to 7 Oct 2026

Stabilizing autoregressive forecasts in chaotic systems via multi-rate latent recurrence

arXiv 2601.14487 ↗First posted 20 Jan 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Long-horizon autoregressive forecasting of chaotic dynamical systems remains challenging due to rapid error amplification and distribution shift: small one-step inaccuracies compound into physically inconsistent rollouts and collapse of large-scale statistics. We introduce MSR-HINE, a hierarchical implicit forecaster that augments multiscale latent priors with multi-rate recurrent modules operating at distinct temporal scales. At each step, coarse-to-fine recurrent states generate latent priors, an implicit one-step predictor refines the state with multiscale latent injections, and a gated fusion with posterior latents enforces scale-consistent updates; a lightweight hidden-state correction further aligns recurrent memories with fused latents. The resulting architecture maintains long-term context on slow manifolds while preserving fast-scale variability, mitigating error accumulation in chaotic rollouts. Across two canonical benchmarks, MSR-HINE yields substantial gains over a U-Net autoregressive baseline: on Kuramoto-Sivashinsky it reduces end-horizon RMSE by 62.8% at H=400 and improves end-horizon ACC by +0.983 (from -0.155 to 0.828), extending the ACC >= 0.5 predictability horizon from 241 to 400 steps; on Lorenz-96 it reduces RMSE by 27.0% at H=100 and improves end horizon ACC by +0.402 (from 0.144 to 0.545), extending the ACC >= 0.5 horizon from 58 to 100 steps.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 56 references that are in the map. First posted on arXiv 20 Jan 2026.

Builds on

34 of 56 references.

References · 34

  1. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  2. 2019DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operatorscited by 1,965
  3. 2020Learning Mesh-Based Simulation with Graph Networkscited by 840
  4. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  5. 2017DGM: A deep learning algorithm for solving partial differential equationscited by 674
  6. 2017The Deep Ritz method: A deep learning-based numerical algorithm for solving variational problemscited by 581
  7. 2017Deep learning for universal linear embeddings of nonlinear dynamicscited by 491
  8. 2021Neural Operator: Learning Maps Between Function Spacescited by 413
  9. 2015A Recurrent Latent Variable Model for Sequential Datacited by 362
  10. 2021Choose a Transformer: Fourier or Galerkincited by 333
  11. 2023Geometry-Informed Neural Operator for Large-Scale 3D PDEscited by 293
  12. 2022U-NO: U-shaped Neural Operatorscited by 219
22 more
  1. 2016Professor Forcing: A New Algorithm for Training Recurrent Networkscited by 214
  2. 2022Towards Multi-spatiotemporal-scale Generalized PDE Modelingcited by 190
  3. 2016Structured Inference Networks for Nonlinear State Space Modelscited by 182
  4. 2019Latent ODEs for Irregularly-Sampled Time Seriescited by 182
  5. 2015Deep Kalman Filterscited by 163
  6. 2018Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networkscited by 155
  7. 2016Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Datacited by 154
  8. 2019Backpropagation Algorithms and Reservoir Computing in Recurrent Neural Networks for the Forecasting of Complex Spatiotemporal Dynamicscited by 132
  9. 2021KalmanNet: Neural Network Aided Kalman Filtering for Partially Known Dynamicscited by 128
  10. 2016Hierarchical Multiscale Recurrent Neural Networkscited by 102
  11. 2016Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequencescited by 98
  12. 2023Operator Learning with Neural Fields: Tackling PDEs on General Geometriescited by 80
  13. 2022Wavelet neural operator: a neural operator for parametric partial differential equationscited by 77
  14. 2020Combining data assimilation and machine learning to emulate a dynamical model from sparse and noisy observations: a case study with the Lorenz 96 modelcited by 72
  15. 2018Hybrid Forecasting of Chaotic Processes: Using Machine Learning in Conjunction with a Knowledge-Based Modelcited by 57
  16. 2023Training neural operators to preserve invariant measures of chaotic attractorscited by 43
  17. 2019ODE2VAE: Deep generative second order ODEs with Bayesian neural networkscited by 36
  18. 2020Learning Variational Data Assimilation Models and Solverscited by 36
  19. 2018Using a reservoir computer to learn chaotic attractors, with applications to chaos synchronisation and cryptographycited by 16
  20. 2023NUNO: A General Framework for Learning Parametric PDEs with Non-Uniform Datacited by 15
  21. 2025Geometry aware inference of steady state PDEs using Equivariant Neural Fields representationscited by 5
  22. 2025Convolutional-neural-operator-based transfer learning for solving PDEscited by 2

Built on by

Not cited yet.