FraudFox: Adaptable Fraud Detection in the Real World

2026-03-13

AI Summary

This work proposes a real-time decision framework for adversarial fraud detection in resource-constrained environments, integrating multi-source risk scores while adhering to operational constraints such as investigation capacity or maximum allowable loss. The framework dynamically updates the weights of individual assessment modules via an extended Kalman filter and incorporates a dynamic importance weighting mechanism to adaptively respond to evolving fraud strategies. Furthermore, it derives a Pareto-optimal decision boundary that enables flexible “what-if” scenario analysis. Deployed in Amazon’s production environment, the system demonstrates significant improvements in detection performance while maintaining scalability, adaptability, and operational effectiveness.

Technology Category

• Multiagent Systems: Adversarial Agents
• Search and Optimization: Adversarial Search
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The proposed method (FraudFox) provides solutions to adversarial attacks in a resource constrained environment. We focus on questions like the following: How suspicious is `Smith', trying to buy \$500 shoes, on Monday 3am? How to merge the risk scores, from a handful of risk-assessment modules (`oracles') in an adversarial environment? More importantly, given historical data (orders, prices, and what-happened afterwards), and business goals/restrictions, which transactions, like the `Smith' transaction above, which ones should we `pass', versus send to human investigators? The business restrictions could be: `at most $x$ investigations are feasible', or `at most \$$y$ lost due to fraud'. These are the two research problems we focus on, in this work. One approach to address the first problem (`oracle-weighting'), is by using Extended Kalman Filters with dynamic importance weights, to automatically and continuously update our weights for each 'oracle'. For the second problem, we show how to derive an optimal decision surface, and how to compute the Pareto optimal set, to allow what-if questions. An important consideration is adaptation: Fraudsters will change their behavior, according to our past decisions; thus, we need to adapt accordingly. The resulting system, \method, is scalable, adaptable to changing fraudster behavior, effective, and already in \textbf{production} at Amazon. FraudFox augments a fraud prevention sub-system and has led to significant performance gains.

Problem

• fraud detection
• adversarial environment
• risk scoring
• resource constraints
• transaction decision

Innovation

• adaptable fraud detection
• oracle-weighting
• Extended Kalman Filter
• Pareto optimal decision
• adversarial environment