Adversarial Robustness in Financial Machine Learning: Defenses, Economic Impact, and Governance Evidence

2025/12/14 by Baviskar, Samruddhi
Computer Science · Business, Management and Accounting · #Adversarial Robustness in Machine Learning #Financial Distress and Bankruptcy Prediction #Explainable Artificial Intelligence (XAI)

paper · pdf · doi:10.48550/arxiv.2512.15780

Abstract

We evaluate adversarial robustness in tabular machine learning models used in financial decision making. Using credit scoring and fraud detection data, we apply gradient based attacks and measure impacts on discrimination, calibration, and financial risk metrics. Results show notable performance degradation under small perturbations and partial recovery through adversarial training.

Citations

• Explaining by Removing: A Unified Framework for Model Explanation
• Adversarial vulnerability for any classifier
• Towards Deep Learning Models Resistant to Adversarial Attacks
• On Calibration of Modern Neural Networks
• Evasion and Hardening of Tree Ensemble Classifiers
• Explaining and Harnessing Adversarial Examples

Related

• Robust Classification of Financial Risk
• Conditional Adversarial Fragility in Financial Machine Learning under Macroeconomic Stress
• PermuteAttack: Counterfactual Explanation of Machine Learning Credit Scorecards
• Foe for Fraud: Transferable Adversarial Attacks in Credit Card Fraud Detection
• Adversarial training for tabular data with attack propagation