Robust Federated Learning with Global Sensitivity Estimation for Financial Risk Management

TL;DR

This paper introduces FRAL-CSE, a federated learning framework for financial risk management that uses global sensitivity estimation and risk measures to improve robustness, scalability, and convergence in decentralized financial systems.

Contribution

The paper proposes a novel FL framework incorporating global sensitivity estimation and curvature-informed updates for enhanced robustness and efficiency in financial risk applications.

Findings

Accelerates convergence compared to existing methods

Enhances robustness against tail risks and extreme scenarios

Effective across heterogeneous financial datasets

Abstract

In decentralized financial systems, robust and efficient Federated Learning (FL) is promising to handle diverse client environments and ensure resilience to systemic risks. We propose Federated Risk-Aware Learning with Central Sensitivity Estimation (FRAL-CSE), an innovative FL framework designed to enhance scalability, stability, and robustness in collaborative financial decision-making. The framework's core innovation lies in a central acceleration mechanism, guided by a quadratic sensitivity-based approximation of global model dynamics. By leveraging local sensitivity information derived from robust risk measurements, FRAL-CSE performs a curvature-informed global update that efficiently incorporates second-order information without requiring repeated local re-evaluations, thereby enhancing training efficiency and improving optimization stability. Additionally, distortion risk measures are embedded into the training objectives to capture tail risks and ensure robustness against extreme scenarios. Extensive experiments validate the effectiveness of FRAL-CSE in accelerating convergence and improving resilience across heterogeneous datasets compared to state-of-the-art baselines.