Implementation of an Asymmetric Adjusted Activation Function for Class Imbalance Credit Scoring

2025-01-21

AI Summary

Severe class imbalance (high imbalance ratio, IR) in credit scoring impairs minority-class (default) discrimination and induces significant prediction bias. To address this, we propose an IR-adaptive asymmetric sigmoid activation function (ASIG), the first to explicitly incorporate the imbalance ratio into activation function design. ASIG dynamically shifts and rescales the sigmoid’s sensitive region, biasing the decision boundary toward the minority class—thereby mitigating positive-sample underlearning at the model’s architectural foundation. Unlike conventional approaches, ASIG requires no resampling or loss reweighting and seamlessly integrates into standard binary-classification neural networks. Extensive evaluation across multiple credit datasets with varying IRs demonstrates that ASIG consistently improves AUC, F1-score, and G-mean. Notably, it maintains robust performance even under extreme imbalance (IR > 100), outperforming state-of-the-art credit scoring models.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Credit scoring is a systematic approach to evaluate a borrower's probability of default (PD) on a bank loan. The data associated with such scenarios are characteristically imbalanced, complicating binary classification owing to the often-underestimated cost of misclassification during the classifier's learning process. Considering the high imbalance ratio (IR) of these datasets, we introduce an innovative yet straightforward optimized activation function by incorporating an IR-dependent asymmetric adjusted factor embedded Sigmoid activation function (ASIG). The embedding of ASIG makes the sensitive margin of the Sigmoid function auto-adjustable, depending on the imbalance nature of the datasets distributed, thereby giving the activation function an asymmetric characteristic that prevents the underrepresentation of the minority class (positive samples) during the classifier's learning process. The experimental results show that the ASIG-embedded-classifier outperforms traditional classifiers on datasets across wide-ranging IRs in the downstream credit-scoring task. The algorithm also shows robustness and stability, even when the IR is ultra-high. Therefore, the algorithm provides a competitive alternative in the financial industry, especially in credit scoring, possessing the ability to effectively process highly imbalanced distribution data.

Problem

• Credit Scoring
• Imbalanced Datasets
• Model Fairness

Innovation

• Asymmetric Adjusted Activation Function
• Imbalanced Data Handling
• Credit Scoring Accuracy