Meta-models for transfer learning in source localisation

TL;DR

This paper introduces a Bayesian multilevel meta-modeling approach for transfer learning in acoustic emission source localisation, capturing inter-task dependencies to predict model hyperparameters for new experiments.

Contribution

It presents a novel hierarchical meta-model framework that encodes inter-experiment knowledge to improve transfer learning in source localisation tasks.

Findings

Effective transfer of hyperparameters to unobserved systems

Meta-model captures inter-task relationships accurately

Improves localisation accuracy in new experiments

Abstract

In practice, non-destructive testing (NDT) procedures tend to consider experiments (and their respective models) as distinct, conducted in isolation and associated with independent data. In contrast, this work looks to capture the interdependencies between acoustic emission (AE) experiments (as meta-models) and then use the resulting functions to predict the model hyperparameters for previously unobserved systems. We utilise a Bayesian multilevel approach (similar to deep Gaussian Processes) where a higher level meta-model captures the inter-task relationships. Our key contribution is how knowledge of the experimental campaign can be encoded between tasks as well as within tasks. We present an example of AE time-of-arrival mapping for source localisation, to illustrate how multilevel models naturally lend themselves to representing aggregate systems in engineering. We constrain the meta-model based on domain knowledge, then use the inter-task functions for transfer learning, predicting hyperparameters for models of previously unobserved experiments (for a specific design).