Countering Overfitting with Counterfactual Examples

TL;DR

This paper introduces CF-Reg, a new regularization method that reduces overfitting by increasing the margin between data points and their counterfactuals, leading to better generalization.

Contribution

We propose CF-Reg, a novel regularization term that explicitly controls overfitting by leveraging counterfactual examples during training.

Findings

CF-Reg outperforms existing regularization techniques across multiple datasets.

Higher overfitting correlates with easier generation of counterfactuals.

Counterfactual regularization improves model generalization.

Abstract

Overfitting is a well-known issue in machine learning that occurs when a model struggles to generalize its predictions to new, unseen data beyond the scope of its training set. Traditional techniques to mitigate overfitting include early stopping, data augmentation, and regularization. In this work, we demonstrate that the degree of overfitting of a trained model is correlated with the ability to generate counterfactual examples. The higher the overfitting, the easier it will be to find a valid counterfactual example for a randomly chosen input data point. Therefore, we introduce CF-Reg, a novel regularization term in the training loss that controls overfitting by ensuring enough margin between each instance and its corresponding counterfactual. Experiments conducted across multiple datasets and models show that our counterfactual regularizer generally outperforms existing regularization techniques.