Fair Multi-party Machine Learning -- a Game Theoretic approach

TL;DR

This paper explores fair data sharing among multiple agents using game theory, analyzing stability, fairness, and profitability of collaborative machine learning models, and proposing new theoretical frameworks and experimental insights.

Contribution

It introduces a novel class of cooperative games for data sharing, with new definitions of stability and fairness, and analyzes their theoretical properties and practical implications.

Findings

Stable and fair outcomes depend on data valuation methods.

Theoretical analysis reveals conditions for optimal and suboptimal sharing.

Experimental results provide insights into data valuation and collaboration incentives.

Abstract

High performance machine learning models have become highly dependent on the availability of large quantity and quality of training data. To achieve this, various central agencies such as the government have suggested for different data providers to pool their data together to learn a unified predictive model, which performs better. However, these providers are usually profit-driven and would only agree to participate inthe data sharing process if the process is deemed both profitable and fair for themselves. Due to the lack of existing literature, it is unclear whether a fair and stable outcome is possible in such data sharing processes. Hence, we wish to investigate the outcomes surrounding these scenarios and study if data providers would even agree to collaborate in the first place. Tapping on cooperative game concepts in Game Theory, we introduce the data sharing process between a group of agents as a new class of cooperative games with modified definition of stability and fairness. Using these new definitions, we then theoretically study the optimal and suboptimal outcomes of such data sharing processes and their sensitivity to perturbation.Through experiments, we present intuitive insights regarding theoretical results analysed in this paper and discuss various ways in which data can be valued reasonably.