Generalization Errors and Learning Curves for Regression with Multi-Task Gaussian Processes

Abstract

We provide some insights into how task correlations in multi-task Gaussian process (GP) regression affect the generalization error and the learning curve. We analyze the asymmetric two-task case, where a secondary task is to help the learning of a primary task. Within this setting, we give bounds on the generalization error and the learning curve of the primary task. Our approach admits intuitive understandings of the multi-task GP by relating it to single-task GPs. For the case of one-dimensional input-space under optimal sampling with data only for the secondary task, the limitations of multi-task GP can be quantified explicitly.

Cite

Text

Chai. "Generalization Errors and Learning Curves for Regression with Multi-Task Gaussian Processes." Neural Information Processing Systems, 2009.

Markdown

[Chai. "Generalization Errors and Learning Curves for Regression with Multi-Task Gaussian Processes." Neural Information Processing Systems, 2009.](https://mlanthology.org/neurips/2009/chai2009neurips-generalization/)

BibTeX

@inproceedings{chai2009neurips-generalization,
  title     = {{Generalization Errors and Learning Curves for Regression with Multi-Task Gaussian Processes}},
  author    = {Chai, Kian M.},
  booktitle = {Neural Information Processing Systems},
  year      = {2009},
  pages     = {279-287},
  url       = {https://mlanthology.org/neurips/2009/chai2009neurips-generalization/}
}