Abstract:We study collaborative learning systems in which the participants are competitors who will defect from the system if they lose revenue by collaborating. As such, we frame the system as a duopoly of competitive firms who are each engaged in training machine-learning models and selling their predictions to a market of consumers. We first examine a fully collaborative scheme in which both firms share their models with each other and show that this leads to a market collapse with the revenues of both firms going to zero. We next show that one-sided collaboration in which only the firm with the lower-quality model shares improves the revenue of both firms. Finally, we propose a more equitable, *defection-free* scheme in which both firms share with each other while losing no revenue, and we show that our algorithm converges to the Nash bargaining solution.

What problem does this paper attempt to address?

This paper discusses a cooperative learning system in a competitive environment, where participants (such as two competing companies) may choose to exit if cooperation leads to a decrease in revenue. The background of the study is two autonomous driving car companies with different datasets, where each trains their own autonomous driving model, but their data cannot fully cover real-world scenarios. The problem addressed in the paper is how to design a cooperative scheme that allows both companies to share models without losing revenue, thereby improving overall performance and encouraging participation. The main contributions of the paper include: 1. Revealing unexpected results of two possible cooperation strategies: when both companies fully share the model, the model quality is maximized but the revenue is reduced to zero; whereas when only the company with lower quality models shares, the quality and revenue of both companies improve. 2. Proposing a non-defection algorithm that allows both companies to contribute models throughout the entire training process without reducing revenue. 3. Proving that the algorithm converges to the Nash bargaining solution in most cases, even if both companies only focus on maximizing their own revenue, thereby achieving the maximization of mutual interests. The paper also discusses related work, including collaborative learning, market competition theory, and how to incentivize cooperation without compromising company profits. Through simulation experiments, the paper demonstrates how the proposed approach functions in practice, with improvements in model quality and increased revenue for both companies, approaching the Nash bargaining solution. In summary, the paper attempts to address how to design a fair cooperation mechanism in a competitive environment, promoting improvements in model performance while ensuring the economic interests of all participants.

Defection-Free Collaboration between Competitors in a Learning System

Privacy-Preserving Collaborative Deep Learning with Unreliable Participants.

Incentivizing Honesty among Competitors in Collaborative Learning and Optimization

Duopoly Business Competition in Cross-Silo Federated Learning

Unravelling in Collaborative Learning

Naive Algorithmic Collusion: When Do Bandit Learners Cooperate and When Do They Compete?

Robust Algorithmic Collusion

Collaborative Machine Learning Markets with Data-Replication-Robust Payments

When Federated Learning Meets Oligopoly Competition: Stability and Model Differentiation

Technical Report: Coopetition in Heterogeneous Cross-Silo Federated Learning

Collaboration Equilibrium in Federated Learning

Algorithmic Collusion: Insights from Deep Learning

Loss aversion fosters coordination among independent reinforcement learners

Algorithmic collusion under competitive design

Joint Training of Deep Ensembles Fails Due to Learner Collusion

On the Conflict of Robustness and Learning in Collaborative Machine Learning

Online Optimization Algorithms in Repeated Price Competition: Equilibrium Learning and Algorithmic Collusion

Learning Collusion in Episodic, Inventory-Constrained Markets

Algorithmic Collusion in Cournot Duopoly Market: Evidence from Experimental Economics

Similarity-based cooperative equilibrium

Uncoupled Learning of Differential Stackelberg Equilibria with Commitments