When is an SHM problem a Multi-Task-Learning problem?
This work is incremental, applying known MTL concepts to the domain of structural health monitoring to potentially improve task performance.
The paper explores how structural health monitoring (SHM) problems can be framed as multi-task learning (MTL) by detailing three mechanisms—natural multiple tasks, using outputs as inputs, and additional loss functions—and provides examples for each.
Multi-task neural networks learn tasks simultaneously to improve individual task performance. There are three mechanisms of multi-task learning (MTL) which are explored here for the context of structural health monitoring (SHM): (i) the natural occurrence of multiple tasks; (ii) using outputs as inputs (both linked to the recent research in population-based SHM (PBSHM)); and, (iii) additional loss functions to provide different insights. Each of these problem settings for MTL is detailed and an example is given.