Low-Cost Learning via Active Data Procurement

Abstract

We design mechanisms for online procurement of data held by strategic agents for machine learning tasks. We study a model in which agents cannot fabricate data, but may lie about their cost of furnishing their data. The challenge is to use past data to actively price future data in order to obtain learning guarantees, even when agents' costs can depend… (More)
DOI: 10.1145/2764468.2764519

4 Figures and Tables

Topics

  • Presentations referencing similar topics