Creator Incentives in Recommender Systems: A Cooperative Game-Theoretic Approach for Stable and Fair Collaboration in Multi-Agent Bandits

TL;DR

This paper models creator incentives in recommender systems as a cooperative game using multi-agent bandits, ensuring stability and fairness through core and Shapley value concepts, with practical payout rules and empirical validation.

Contribution

It introduces a cooperative game-theoretic framework for incentivizing content creators in recommendation platforms, analyzing stability and fairness for homogeneous and heterogeneous agents.

Findings

Convexity of the TU game for homogeneous agents ensures a non-empty core.

The proposed payout rule satisfies most Shapley axioms and lies in the core.

Empirical results show alignment and divergence of payouts with Shapley fairness in real data.

Abstract

User interactions in online recommendation platforms create interdependencies among content creators: feedback on one creator's content influences the system's learning and, in turn, the exposure of other creators' contents. To analyze incentives in such settings, we model collaboration as a multi-agent stochastic linear bandit problem with a transferable utility (TU) cooperative game formulation, where a coalition's value equals the negative sum of its members' cumulative regrets. We show that, for identical (homogenous) agents with fixed action sets, the induced TU game is convex under mild algorithmic conditions, implying a non-empty core that contains the Shapley value and ensures both stability and fairness. For heterogeneous agents, the game still admits a non-empty core, though convexity and Shapley value core-membership are no longer guaranteed. To address this, we propose a simple regret-based payout rule that satisfies three out of the four Shapley axioms and also lies in the core. Experiments on MovieLens-100k dataset illustrate when the empirical payout aligns with -- and diverges from -- the Shapley fairness across different settings and algorithms.