Mona Anjali Hana Menno Bire

Topology-Aware Reinforcement Learning for Adaptive Credit Card Fraud Detection: Integrating Multidimensional Rebalancing with Deep Q-Networks

Background Credit card fraud detection faces extreme class imbalance, concept drift, and topological overlap. Standard Deep Q-Networks (DQN) suffer catastrophic performance degradation (e.g., 26.8% precision) due to noisy reward signals from overlapping decision boundaries. Methods We propose the Adaptive Multidimensio...