RefreshNet: Learning multiscale dynamics through…

In brief

Forecasts that step forward one small interval at a time build up error over long horizons. RefreshNet stacks models that work at different time scales and lets the slower ones periodically correct the faster ones.

Abstract

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.

Cite this paper

@article{farooq2024refreshnet,
  title   = {RefreshNet: Learning multiscale dynamics through hierarchical refreshing},
  author  = {Farooq, Junaid and Rafiq, Danish and Vlachas, Pantelis R. and Bazaz, Mohammad Abid},
  journal = {Nonlinear Dynamics},
  year    = {2024},
  volume  = {112(16)},
  pages   = {14479--14496},
  doi     = {10.1007/s11071-024-09813-3},
}

Topic: Scientific machine learning

All publications