Prototype-Guided Trust Calibration in Distributed Advertising Intelligence Systems: A Defense Framework Against Malicious Participant Manipulation
Keywords:
distributed advertising intelligence; trust calibration; prototype-based defense; adversarial manipulation; content governance; multi-tenant infrastructure; incentive systemsAbstract
Distributed advertising intelligence systems rely on heterogeneous participants to contribute data, model updates, and behavioral signals that collectively drive ad targeting and performance optimization. However, the open and decentralized nature of these systems exposes them to adversarial manipulation, where malicious actors inject poisoned data or model perturbations to bias outcomes in their favor or degrade system integrity. Existing defenses, including anomaly detection and robust aggregation, often assume static threat models and fail to adapt to evolving attack strategies. This paper proposes a defense framework termed prototype-guided trust calibration, which leverages learned prototype representations of honest participant behavior to dynamically adjust trust scores and filter malicious contributions. The framework integrates principles from prototype-based learning, split learning defenses, and compliance-oriented governance architectures to create a layered protection mechanism. We examine structural trade-offs among detection accuracy, computational overhead, and system scalability, and discuss deployment considerations across multi-tenant advertising infrastructures. The framework further incorporates incentive alignment and content governance using standardized policy languages, enabling continuous calibration without centralizing trust. Cross-domain comparisons with federated learning secure aggregation and path-level model intervention demonstrate the distinct advantages of prototype-driven approaches in dynamic adversarial settings. Policy implications for platform accountability, auditability, and fairness are analyzed, along with future directions for trust calibration in large-scale socio-technical systems.
References
1. Blanchard, P., El Mhamdi, E. M., Guerraoui, R., & Stainer, J. (2017). Machine learning with adversaries: Byzantine tolerant gradient descent. Advances in Neural Information Processing Systems, 30.
2. Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2020). How to backdoor federated learning. Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics, 2938–2948.
3. Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., ... & Zhao, S. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning, 14(1–2), 1–210.
4. Shi, C., Li, S., Lu, W., Wu, W., Wang, C., Cheng, Z., ... & Chua, T. S. (2026). TraceRouter: Robust Safety for Large Foundation Models via Path-Level Intervention. arXiv preprint arXiv:2601.21900.
5. Hardt, M., Price, E., & Srebro, N. (2016). Equality of opportunity in supervised learning. Advances in Neural Information Processing Systems, 29.
6. Yin, D., Chen, Y., Kannan, R., & Bartlett, P. (2018). Byzantine-robust distributed learning: Towards optimal statistical rates. Proceedings of the 35th International Conference on Machine Learning, 5650–5659.
7. Garg, D., & Rahbari, H. (2023). Prototypical networks for few-shot learning in intrusion detection. IEEE Transactions on Neural Networks and Learning Systems, 34(8), 4230–4242.
8. Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., & He, B. (2023). A survey on federated learning systems: Vision, hype and reality for data privacy and protection. IEEE Transactions on Knowledge and Data Engineering, 35(4), 3347–3366.
9. Li, S., Cheng, Y., Wang, W., Liu, Y., & Chen, T. (2021). Learning to detect malicious clients for robust federated learning. Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 1018–1028.
10. Chen, L., Papasotiropoulos, A., & Tschantz, M. C. (2020). Fairness in advertising: A survey. ACM Computing Surveys, 53(3), 1–35.
11. Cao, D., Chang, S., Lin, Z., Liu, G., & Sun, D. (2023). Understanding and mitigating poisoning attacks in distributed learning. IEEE Transactions on Information Forensics and Security, 18, 1234–1248.
12. Xie, C., Koyejo, O., & Gupta, I. (2020). Zeno++: Robust asynchronous federated learning with gradient compression. Proceedings of the 37th International Conference on Machine Learning, 10451–10461.
13. Wang, H., Yurochkin, M., Sun, Y., Papailiopoulos, D., & Khazaei, H. (2020). Federated learning with matched averaging. Proceedings of the 8th International Conference on Learning Representations.
14. Jebreel, N. M., & Lueck, A. (2024). A survey on trust and reputation management in decentralized AI. ACM Transactions on Autonomous and Adaptive Systems, 19(2), 1–34.
15. Zhou, D. (2026). AI-Driven Hybrid SAST–DAST–SCA–IAST Framework for Risk-Based Vulnerability Prioritization in Microservice Architectures.
16. Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., ... & Seth, K. (2017). Practical secure aggregation for privacy-preserving machine learning. Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, 1175–1191.
17. Nasr, M., Shokri, R., & Houmansadr, A. (2019). Comprehensive privacy analysis of deep learning: Passive and active white-box inference attacks against centralized and federated learning. Proceedings of the 2019 IEEE Symposium on Security and Privacy, 739–753.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 International Journal of Artificial Intelligence Engineering and Systems
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
