The Effects of Data Imbalance Under a Federated Learning Approach for Credit Risk Forecasting

TL;DR

This paper investigates how federated learning can be applied to credit risk forecasting, especially under data imbalance conditions, demonstrating its advantages for smaller clients and highlighting challenges for dominant clients.

Contribution

It evaluates federated learning's effectiveness in credit risk assessment across various models and datasets, emphasizing its benefits for non-dominant clients in imbalanced data scenarios.

Findings

Federated models outperform local models on small, non-dominant clients.

Data imbalance amplifies federated learning benefits, with an average 17.92% performance boost.

Federated learning may not always outperform local models for clients with abundant data.

Abstract

Credit risk forecasting plays a crucial role for commercial banks and other financial institutions in granting loans to customers and minimise the potential loss. However, traditional machine learning methods require the sharing of sensitive client information with an external server to build a global model, potentially posing a risk of security threats and privacy leakage. A newly developed privacy-preserving distributed machine learning technique known as Federated Learning (FL) allows the training of a global model without the necessity of accessing private local data directly. This investigation examined the feasibility of federated learning in credit risk assessment and showed the effects of data imbalance on model performance. Two neural network architectures, Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM), and one tree ensemble architecture, Extreme Gradient Boosting (XGBoost), were explored across three different datasets under various scenarios involving different numbers of clients and data distribution configurations. We demonstrate that federated models consistently outperform local models on non-dominant clients with smaller datasets. This trend is especially pronounced in highly imbalanced data scenarios, yielding a remarkable average improvement of 17.92% in model performance. However, for dominant clients (clients with more data), federated models may not exhibit superior performance, suggesting the need for special incentives for this type of clients to encourage their participation.