Paper Type

ERF

Abstract

Cybersecurity attacks impose operational, financial, and reputational harm to organizations, yet information sharing remains limited due to competitive and disclosure concerns. We propose Federated Learning (FL) as a solution, using a game-theoretic model, in which competing firms invest in cybersecurity and participate in an FL system that improves attack detection without sharing raw data. By linking detection effectiveness to both collective participation and individual contribution quality, the model embeds incentive alignment directly into the learning process. Our model can reduce free-riding and support information sharing even in competitive markets. Our findings show that carefully designed technological mechanisms can support mutually beneficial cybersecurity collaboration.

Paper Number

1494

Comments

SIG SEC

Download

Share

COinS
 
Aug 15th, 12:00 AM

Contribution-based Federated Learning in Cybersecurity

Cybersecurity attacks impose operational, financial, and reputational harm to organizations, yet information sharing remains limited due to competitive and disclosure concerns. We propose Federated Learning (FL) as a solution, using a game-theoretic model, in which competing firms invest in cybersecurity and participate in an FL system that improves attack detection without sharing raw data. By linking detection effectiveness to both collective participation and individual contribution quality, the model embeds incentive alignment directly into the learning process. Our model can reduce free-riding and support information sharing even in competitive markets. Our findings show that carefully designed technological mechanisms can support mutually beneficial cybersecurity collaboration.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.