System Identification with Time-Aware Neural Sequence Models

TL;DR

This paper introduces a time-aware extension of neural sequence models, including GRUs, to improve dynamical system identification from unevenly sampled continuous data, enabling better modeling of temporal dynamics.

Contribution

It proposes a novel time-aware and stationary extension of neural sequence models that naturally handle variable sampling intervals in system identification tasks.

Findings

Effective handling of uneven sampling in system identification

Demonstrated on industrial input/output data

Improved modeling of higher-order temporal behavior

Abstract

Established recurrent neural networks are well-suited to solve a wide variety of prediction tasks involving discrete sequences. However, they do not perform as well in the task of dynamical system identification, when dealing with observations from continuous variables that are unevenly sampled in time, for example due to missing observations. We show how such neural sequence models can be adapted to deal with variable step sizes in a natural way. In particular, we introduce a time-aware and stationary extension of existing models (including the Gated Recurrent Unit) that allows them to deal with unevenly sampled system observations by adapting to the observation times, while facilitating higher-order temporal behavior. We discuss the properties and demonstrate the validity of the proposed approach, based on samples from two industrial input/output processes.