English

Explaining How Deep Neural Networks Forget by Deep Visualization

Machine Learning 2022-07-26 v3 Computer Vision and Pattern Recognition

Abstract

Explaining the behaviors of deep neural networks, usually considered as black boxes, is critical especially when they are now being adopted over diverse aspects of human life. Taking the advantages of interpretable machine learning (interpretable ML), this paper proposes a novel tool called Catastrophic Forgetting Dissector (or CFD) to explain catastrophic forgetting in continual learning settings. We also introduce a new method called Critical Freezing based on the observations of our tool. Experiments on ResNet articulate how catastrophic forgetting happens, particularly showing which components of this famous network are forgetting. Our new continual learning algorithm defeats various recent techniques by a significant margin, proving the capability of the investigation. Critical freezing not only attacks catastrophic forgetting but also exposes explainability.

Keywords

Cite

@article{arxiv.2005.01004,
  title  = {Explaining How Deep Neural Networks Forget by Deep Visualization},
  author = {Giang Nguyen and Shuan Chen and Tae Joon Jun and Daeyoung Kim},
  journal= {arXiv preprint arXiv:2005.01004},
  year   = {2022}
}

Comments

12 pages, 4 figures, 1 table. arXiv admin note: substantial text overlap with arXiv:2001.01578. Keywords: XAI, Catastrophic Forgetting, Attribution map, Continual Learning and Regularization

R2 v1 2026-06-23T15:16:11.217Z