paper-with-me

홈 › Papers

FSC-Net: Fast-Slow Consolidation Networks for Continual Learning

2025-11-12 · Mohamed El Gorrim arxiv

Continual learning remains challenging due to catastrophic forgetting, where neural networks lose previously acquired knowledge when learning new tasks. Inspired by memory consolidation in neuroscience, we propose FSC-Net (Fast-Slow Consolidation Networks), a dual-network architecture that separates rapid task learning from gradual knowledge consolidation. Our method employs a fast network (NN1) for immediate adaptation to new tasks and a slow network (NN2) that consolidates knowledge through distillation and replay. Within the family of MLP-based NN1 variants we evaluated, consolidation effectiveness is driven more by methodology than architectural embellishments -- a simple MLP outperforms more complex similarity-gated variants by 1.2pp. Through systematic hyperparameter analysis, we observed empirically that pure replay without distillation during consolidation achieves superior performance, consistent with the hypothesis that distillation from the fast network introduces recency bias. On Split-MNIST (30 seeds), FSC-Net achieves 91.71% +/- 0.62% retention accuracy, a +4.27pp gain over the fast network alone (87.43% +/- 1.27%, paired t=23.585, p < 1e-10). On Split-CIFAR-10 (5 seeds), our method achieves 33.31% +/- 0.38% retention with an +8.20pp gain over the fast network alone (25.11% +/- 1.61%, paired t=9.75, p < 1e-3), demonstrating +8.20pp gain, though absolute performance (33.31%) remains modest and below random expectation, highlighting need for stronger backbones. Our results provide empirical evidence that the dual-timescale consolidation mechanism, rather than architectural complexity, is central to mitigating catastrophic forgetting in this setting.

📄 PDF Abstract BibTeX arXiv:2511.11707

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Similar Papers 제목 키워드 기반

Active perception and disentangled representations allow continual, episodic zero and few-shot learning

2026-02-22 · David Rawlinson, Gideon Kowadlo arxiv

Generalization is often regarded as an essential property of machine learning systems. However, perhaps not every component of a system needs to generalize. Training models for generalization typically produces entangled…

Continual LearningFew-Shot Learning

Few-shot Continual Learning: a Brain-inspired Approach

2021-04-19 · Liyuan Wang, Qian Li, Yi Zhong, Jun Zhu

It is an important yet challenging setting to continually learn new tasks from a few examples. Although numerous efforts have been devoted to either continual learning or few-shot learning, little work has considered thi…

Continual LearningFew-Shot Learning

Balancing Plasticity and Stability with Fast and Slow Successor Features

2026-05-25 · Raymond Chua, Doina Precup, Blake Richards arxiv

A hallmark of intelligence is the ability to adapt in non-stationary environments, yet deep Reinforcement Learning (RL) agents often struggle in such settings. Prior studies introduce non-stationarity through abrupt shif…

Reinforcement LearningContinual Learning

Differentiable Hebbian Consolidation for Continual Learning

2019-09-25 · Vithursan Thangarasa, Thomas Miconi, Graham W. Taylor

Continual learning is the problem of sequentially learning new tasks or knowledge while protecting previously acquired knowledge. However, catastrophic forgetting poses a grand challenge for neural networks performing su…

Continual LearningPermuted-MNISTSplit-MNIST

Efficient Continual Learning in Language Models via Thalamically Routed Cortical Columns

2026-02-25 · Afshin Khadangi arxiv

Large language models deployed in the wild must adapt to evolving data, user behavior, and task mixtures without erasing previously acquired capabilities. In practice, this remains difficult: sequential updates induce ca…

Continual Learning