paper-with-me

홈 › Papers

Why Do Neural Networks Forget: A Study of Collapse in Continual Learning

2026-03-04 · Yunqin Zhu, Jun Jin arxiv

Catastrophic forgetting is a major problem in continual learning, and lots of approaches arise to reduce it. However, most of them are evaluated through task accuracy, which ignores the internal model structure. Recent research suggests that structural collapse leads to loss of plasticity, as evidenced by changes in effective rank (eRank). This indicates a link to forgetting, since the networks lose the ability to expand their feature space to learn new tasks, which forces the network to overwrite existing representations. Therefore, in this study, we investigate the correlation between forgetting and collapse through the measurement of both weight and activation eRank. To be more specific, we evaluated four architectures, including MLP, ConvGRU, ResNet-18, and Bi-ConvGRU, in the split MNIST and Split CIFAR-100 benchmarks. Those models are trained through the SGD, Learning-without-Forgetting (LwF), and Experience Replay (ER) strategies separately. The results demonstrate that forgetting and collapse are strongly related, and different continual learning strategies help models preserve both capacity and performance in different efficiency.

📄 PDF Abstract BibTeX arXiv:2603.04580

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Similar Papers 제목 키워드 기반

SAMix: Calibrated and Accurate Continual Learning via Sphere-Adaptive Mixup and Neural Collapse

2025-10-17 · Trung-Anh Dang, Vincent Nguyen, Ngoc-Son Vu, Christel Vrain arxiv

While most continual learning methods focus on mitigating forgetting and improving accuracy, they often overlook the critical aspect of network calibration, despite its importance. Neural collapse, a phenomenon where las…

Continual Learning

On Catastrophic Forgetting and Mode Collapse in Generative Adversarial Networks

2018-07-11 · Hoang Thanh-Tung, Truyen Tran

In this paper, we show that Generative Adversarial Networks (GANs) suffer from catastrophic forgetting even when they are trained to approximate a single target distribution. We show that GAN training is a continual lear…

Continual Learning

Mitigating Mode Collapse by Sidestepping Catastrophic Forgetting

2021-01-01 · Karttikeya Mangalam, Rohin Garg, Jathushan Rajasegaran, Taesung Park

Generative Adversarial Networks (GANs) are a class of generative models used for various applications, but they have been known to suffer from the mode collapse problem, in which some modes of the target distribution are…

Continual Learning

Overcoming Mode Collapse with Adaptive Multi Adversarial Training

2021-12-29 · Karttikeya Mangalam, Rohin Garg

Generative Adversarial Networks (GANs) are a class of generative models used for various applications, but they have been known to suffer from the mode collapse problem, in which some modes of the target distribution are…

Continual Learning

Towards Robustness and Diversity: Continual Learning in Dialog Generation with Text-Mixup and Batch Nuclear-Norm Maximization

2024-03-16 · Zihan Wang, Jiayu Xiao, Mengxiang Li, Zhongjiang He 외

In our dynamic world where data arrives in a continuous stream, continual learning enables us to incrementally add new tasks/domains without the need to retrain from scratch. A major challenge in continual learning of la…

Continual LearningData AugmentationDiversityLanguage Modeling+1