On the Value of Target Data in Transfer Learning
Machine Learning
2020-02-13 v1 Machine Learning
Abstract
We aim to understand the value of additional labeled or unlabeled target data in transfer learning, for any given amount of source data; this is motivated by practical questions around minimizing sampling costs, whereby, target data is usually harder or costlier to acquire than source data, but can yield better accuracy. To this aim, we establish the first minimax-rates in terms of both source and target sample sizes, and show that performance limits are captured by new notions of discrepancy between source and target, which we refer to as transfer exponents.
Cite
@article{arxiv.2002.04747,
title = {On the Value of Target Data in Transfer Learning},
author = {Steve Hanneke and Samory Kpotufe},
journal= {arXiv preprint arXiv:2002.04747},
year = {2020}
}