Meta-Learning Reliable Priors in the Function Space
Abstract
When data are scarce meta-learning can improve a learner's accuracy by harnessing previous experience from related learning tasks. However, existing methods have unreliable uncertainty estimates which are often overconfident. Addressing these shortcomings, we introduce a novel meta-learning framework, called F-PACOH, that treats meta-learned priors as stochastic processes and performs meta-level regularization directly in the function space. This allows us to directly steer the probabilistic predictions of the meta-learner towards high epistemic uncertainty in regions of insufficient meta-training data and, thus, obtain well-calibrated uncertainty estimates. Finally, we showcase how our approach can be integrated with sequential decision making, where reliable uncertainty quantification is imperative. In our benchmark study on meta-learning for Bayesian Optimization (BO), F-PACOH significantly outperforms all other meta-learners and standard baselines.
Cite
@article{arxiv.2106.03195,
title = {Meta-Learning Reliable Priors in the Function Space},
author = {Jonas Rothfuss and Dominique Heyn and Jinfan Chen and Andreas Krause},
journal= {arXiv preprint arXiv:2106.03195},
year = {2022}
}
Comments
In Advances of Neural Information Processing Systems (NeurIPS) 2021