Explaining Model Overfitting in CNNs via GMM Clustering
This work addresses the challenge of interpreting CNN behavior for researchers and practitioners, offering a novel perspective on overfitting, though it appears incremental as it builds on existing clustering methods.
The study tackled the problem of opaque decision-making in CNNs by using GMM clustering to identify anomaly filters linked to outlier samples, providing quantitative metrics and hypotheses to explain model overfitting without modifications to architectures like AlexNet and LeNet-5.
Convolutional Neural Networks (CNNs) have demonstrated remarkable prowess in the field of computer vision. However, their opaque decision-making processes pose significant challenges for practical applications. In this study, we provide quantitative metrics for assessing CNN filters by clustering the feature maps corresponding to individual filters in the model via Gaussian Mixture Model (GMM). By analyzing the clustering results, we screen out some anomaly filters associated with outlier samples. We further analyze the relationship between the anomaly filters and model overfitting, proposing three hypotheses. This method is universally applicable across diverse CNN architectures without modifications, as evidenced by its successful application to models like AlexNet and LeNet-5. We present three meticulously designed experiments demonstrating our hypotheses from the perspectives of model behavior, dataset characteristics, and filter impacts. Through this work, we offer a novel perspective for evaluating the CNN performance and gain new insights into the operational behavior of model overfitting.