Privacy-Aware Data Acquisition under Data Similarity in Regression Markets
This work addresses privacy and data similarity challenges in decentralized data markets for applications like prediction and learning, representing an incremental improvement over prior approaches.
The paper tackles the problem of designing data markets that account for data similarity and privacy preferences, proposing a query-response protocol with local differential privacy and analyzing strategic interactions as a Stackelberg game, with numerical evaluation showing how data similarity affects market participation and traded data value.
Data markets facilitate decentralized data exchange for applications such as prediction, learning, or inference. The design of these markets is challenged by varying privacy preferences as well as data similarity among data owners. Related works have often overlooked how data similarity impacts pricing and data value through statistical information leakage. We demonstrate that data similarity and privacy preferences are integral to market design and propose a query-response protocol using local differential privacy for a two-party data acquisition mechanism. In our regression data market model, we analyze strategic interactions between privacy-aware owners and the learner as a Stackelberg game over the asked price and privacy factor. Finally, we numerically evaluate how data similarity affects market participation and traded data value.