Motivated by the need for efficient and personalized learning in mobile health, we investigate the problem of online kernel selection for Gaussian Process regression in the multi-task setting. We propose a novel generative process on the kernel composition for this purpose. Our method demonstrates that trajectories of kernel evolutions can be transferred between users to improve learning and that the kernels themselves are meaningful for an mHealth prediction goal.
@article{arxiv.2107.09949,
title = {Online structural kernel selection for mobile health},
author = {Eura Shin and Pedja Klasnja and Susan Murphy and Finale Doshi-Velez},
journal= {arXiv preprint arXiv:2107.09949},
year = {2021}
}