Meta-Learning of Neural State-Space Models Using Data From Similar Systems

TL;DR

This paper introduces a meta-learning approach for neural state-space models that leverages data from similar systems to improve modeling accuracy and reduce online adaptation complexity.

Contribution

It applies model-agnostic meta-learning to neural state-space models, enabling rapid adaptation with limited data from the target system.

Findings

Meta-learning improves accuracy over supervised and transfer learning.

Few adaptation steps suffice for accurate modeling.

Selective layer adaptation reduces complexity without sacrificing performance.

Abstract

Deep neural state-space models (SSMs) provide a powerful tool for modeling dynamical systems solely using operational data. Typically, neural SSMs are trained using data collected from the actual system under consideration, despite the likely existence of operational data from similar systems which have previously been deployed in the field. In this paper, we propose the use of model-agnostic meta-learning (MAML) for constructing deep encoder network-based SSMs, by leveraging a combination of archived data from similar systems (used to meta-train offline) and limited data from the actual system (used for rapid online adaptation). We demonstrate using a numerical example that meta-learning can result in more accurate neural SSM models than supervised- or transfer-learning, despite few adaptation steps and limited online data. Additionally, we show that by carefully partitioning and adapting the encoder layers while fixing the state-transition operator, we can achieve comparable performance to MAML while reducing online adaptation complexity.