Empirically Measuring Transfer Distance for System Design and Operation

TL;DR

This paper introduces a formal measure called transfer distance to empirically evaluate and improve the transferability of models across different systems, aiding in system design and operation under data limitations.

Contribution

It defines transfer distance and demonstrates its application in designing prognostic models and predicting performance, addressing data scarcity in transfer learning.

Findings

Transfer distance quantifies model transferability.

Methodology improves prognostic model design.

Predicts operational performance effectively.

Abstract

Classical machine learning approaches are sensitive to non-stationarity. Transfer learning can address non-stationarity by sharing knowledge from one system to another, however, in areas like machine prognostics and defense, data is fundamentally limited. Therefore, transfer learning algorithms have little, if any, examples from which to learn. Herein, we suggest that these constraints on algorithmic learning can be addressed by systems engineering. We formally define transfer distance in general terms and demonstrate its use in empirically quantifying the transferability of models. We consider the use of transfer distance in the design of machine rebuild procedures to allow for transferable prognostic models. We also consider the use of transfer distance in predicting operational performance in computer vision. Practitioners can use the presented methodology to design and operate systems with consideration for the learning theoretic challenges faced by component learning systems.