Cost-Aware Learning for Improved Identifiability with Multiple Experiments
Abstract
We analyze the sample complexity of learning from multiple experiments where the experimenter has a total budget for obtaining samples. In this problem, the learner should choose a hypothesis that performs well with respect to multiple experiments, and their related data distributions. Each collected sample is associated with a cost which depends on the particular experiments. In our setup, a learner performs experiments, while incurring a total cost . We first show that learning from multiple experiments allows to improve identifiability. Additionally, by using a Rademacher complexity approach, we show that the gap between the training and generalization error is . We also provide some examples for linear prediction, two-layer neural networks and kernel methods.
Cite
@article{arxiv.1802.04350,
title = {Cost-Aware Learning for Improved Identifiability with Multiple Experiments},
author = {Longyun Guo and Jean Honorio and John Morgan},
journal= {arXiv preprint arXiv:1802.04350},
year = {2019}
}
Comments
17 pages, 4 figures