Analysis of Wide and Deep Echo State Networks for Multiscale Spatiotemporal Time Series Forecasting

TL;DR

This paper investigates the performance of deep echo state networks in multiscale spatiotemporal forecasting, focusing on neuron partitioning and parallel reservoir pathways to enhance understanding of their capabilities.

Contribution

It provides a comprehensive analysis of deep reservoir computing architectures, emphasizing neuron partitioning and parallel pathways for improved multiscale time series forecasting.

Findings

Partitioning neurons affects forecasting accuracy.

Parallel reservoir pathways improve multiscale modeling.

Deep reservoirs outperform shallow counterparts in complex tasks.

Abstract

Echo state networks are computationally lightweight reservoir models inspired by the random projections observed in cortical circuitry. As interest in reservoir computing has grown, networks have become deeper and more intricate. While these networks are increasingly applied to nontrivial forecasting tasks, there is a need for comprehensive performance analysis of deep reservoirs. In this work, we study the influence of partitioning neurons given a budget and the effect of parallel reservoir pathways across different datasets exhibiting multi-scale and nonlinear dynamics.