Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis
Abstract
This paper addresses the source localization problem by introducing RoLocMe, a multi-agent reinforcement learning system that integrates SkipNet - a skip-connection-based RSS estimation model - with parallel Q-learning. SkipNet predicts RSS propagation of the entire search region, enabling agents to explore efficiently. The agents leverage dueling DQN, value decomposition, and λ-returns to learn cooperative policies. RoLocMe converges faster and achieves at least 20% higher success rates than existing methods in dense and sparse reward settings. A drop-one ablation study confirms each component’s importance and RoLocMe’s effectiveness for larger teams.
Cite
Text
Mayo et al. "Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis." International Joint Conference on Artificial Intelligence, 2024. doi:10.24963/ijcai.2024/18Markdown
[Mayo et al. "Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis." International Joint Conference on Artificial Intelligence, 2024.](https://mlanthology.org/ijcai/2024/mayo2024ijcai-fraud/) doi:10.24963/ijcai.2024/18BibTeX
@inproceedings{mayo2024ijcai-fraud,
title = {{Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis}},
author = {Mayo, Katherine and Grabill, Nicholas and Wellman, Michael P.},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2024},
pages = {157-165},
doi = {10.24963/ijcai.2024/18},
url = {https://mlanthology.org/ijcai/2024/mayo2024ijcai-fraud/}
}