Work

Forecasting Using Reservoir Computing: The Role of Generalized Synchronization

Year: 2021

Type: preprint

Abstract: Reservoir computers (RC) are a form of recurrent neural network (RNN) used for forecasting time series data. As with all RNNs, selecting the hyperparameters presents a challenge when training on new inputs. We present a method based on generalized synchronization (GS) that gives direction in designing and evaluating the architecture and hyperparameters of a RC. The 'auxiliary method' for detecting GS provides a pre-training test that guides hyperparameter selection. Furthermore, we provide a metric for a "well trained" RC using the reproduction of the input system's Lyapunov exponents.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2102.08930

Authors: Platt, Jason A., Wong, Adrian, Clark, Randall, Penny, Stephen G., Abarbanel, Henry D. I.

Institutions: University of California San Diego, National Oceanic and Atmospheric Administration

Language: English

Topic: Neural Networks and Reservoir Computing

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Oceanic and Atmospheric Administration, Office of Naval Research

Awards: G3882977533, G4017054702, G5330321839, G5665182885, G5972442391, G7194740956, G8006577219, G8876996369

OpenAlex ID: W3130744220

Cites: 34

Cited by: 5

Related to: 20