Discovery of slow variables in a class of multiscale stochastic systems via neural networks
arXiv:2104.13911 · doi:10.1007/s00332-022-09808-7
Abstract
Finding a reduction of complex, high-dimensional dynamics to its essential, low-dimensional "heart" remains a challenging yet necessary prerequisite for designing efficient numerical approaches. Machine learning methods have the potential to provide a general framework to automatically discover such representations. In this paper, we consider multiscale stochastic systems with local slow-fast time scale separation and propose a new method to encode in an artificial neural network a map that extracts the slow representation from the system. The architecture of the network consists of an encoder-decoder pair that we train in a supervised manner to learn the appropriate low-dimensional embedding in the bottleneck layer. We test the method on a number of examples that illustrate the ability to discover a correct slow representation. Moreover, we provide an error measure to assess the quality of the embedding and demonstrate that pruning the network can pinpoint an essential coordinates of the system to build the slow representation.
arXiv comment26 pages, 15 figures
References in corpus (11)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Data-driven discovery of coordinates and governing equations
- The State of Sparsity in Deep Neural Networks
- Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
- Identification of simple reaction coordinates from complex dynamics
- Normal form transforms separate slow and fast modes in stochastic dynamical systems
- Effective dynamics using conditional expectations
- Geometric singular perturbation theory for stochastic differential equations
- Transition manifolds of complex metastable systems: Theory and data-driven computation of effective dynamics
- A micro-macro acceleration method for the Monte Carlo simulation of stochastic differential equations
- Multiscale modeling, discretization, and algorithms: a survey in biomechanics