RefreshNet: Learning Multiscale Dynamics through Hierarchical Refreshing
Not on the map. Why?
Abstract (arXiv)
Forecasting complex system dynamics, particularly for long-term predictions, is persistently hindered by error accumulation and computational burdens. This study presents RefreshNet, a multiscale framework developed to overcome these challenges, delivering an unprecedented balance between computational efficiency and predictive accuracy. RefreshNet incorporates convolutional autoencoders to identify a reduced order latent space capturing essential features of the dynamics, and strategically employs multiple recurrent neural network (RNN) blocks operating at varying temporal resolutions within the latent space, thus allowing the capture of latent dynamics at multiple temporal scales. The unique "refreshing" mechanism in RefreshNet allows coarser blocks to reset inputs of finer blocks, effectively controlling and alleviating error accumulation. This design demonstrates superiority over existing techniques regarding computational efficiency and predictive accuracy, especially in long-term forecasting. The framework is validated using three benchmark applications: the FitzHugh-Nagumo system, the Reaction-Diffusion equation, and Kuramoto-Sivashinsky dynamics. RefreshNet significantly outperforms state-of-the-art methods in long-term forecasting accuracy and speed, marking a significant advancement in modeling complex systems and opening new avenues in understanding and predicting their behavior.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 23 of its 73 references that are in the map. First posted on arXiv 24 Jan 2024.
Builds on
23 of 73 references.
References · 23
- 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
- 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,004
- 2013Speech Recognition with Deep Recurrent Neural Networkscited by 666
- 2018Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Networkcited by 415
- 2021A Survey of Transformerscited by 399
- 2015LSTM: A Search Space Odysseycited by 315
- 2016Exploring the Limits of Language Modelingcited by 172
- 2018Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networkscited by 155
- 2017Linearly-Recurrent Autoencoder Networks for Learning Dynamicscited by 142
- 2019Backpropagation Algorithms and Reservoir Computing in Recurrent Neural Networks for the Forecasting of Complex Spatiotemporal Dynamicscited by 132
- 2019AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networkscited by 112
- 2020Multiscale Simulations of Complex Systems by Learning their Effective Dynamicscited by 86
11 more
- 2023Generalized Teacher Forcing for Learning Chaotic Dynamicscited by 55
- 2021Data-driven discovery of intrinsic dynamicscited by 45
- 2019Deep learning to discover and predict dynamics on an inertial manifoldcited by 42
- 2022Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systemscited by 34
- 2015Synthesis of recurrent neural networks for dynamical system simulationcited by 16
- 2021Machine Learning Approach to Model Order Reduction of Nonlinear Systems via Autoencoder and LSTM Networkscited by 16
- 2022Deep autoencoders for physics-constrained data-driven nonlinear materials modelingcited by 13
- 2022Flipped Classroom: Effective Teaching for Time Series Forecastingcited by 11
- 2023Learning from Predictions: Fusing Training and Autoregressive Inference for Long-Term Spatiotemporal Forecastscited by 11
- 2020Multiresolution Convolutional Autoencoderscited by 10
- 2021A spatio-temporal LSTM model to forecast across multiple temporal and spatial scalescited by 6
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.