paper-with-me

홈 › Papers

SingularClip: Preventing Spectral Collapse to Maintain Plasticity in Continual and Reinforcement Learning

2026-08-18 · Tyler Kastner, Nimrod De La Vega, Amir-massoud Farahmand arxiv

Neural networks trained on nonstationary tasks frequently lose the ability to fit new targets, a phenomenon referred to as loss of plasticity. We identify a novel source of plasticity loss due to the growing anisotropy of weight matrices' singular values during training, and analyze this phenomenon both empirically and theoretically. To mitigate this issue, we introduce SingularClip, a procedure that periodically clips the singular values of all weight matrices. We show that SingularClip performs strongly against baselines across a range of tasks in both continual supervised learning and deep reinforcement learning.

📄 PDF Abstract BibTeX arXiv:2608.18319

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Spectral Collapse Drives Loss of Plasticity in Deep Continual Learning

2025-09-26 · Arjun Prakash, Naicheng He, Kaicheng Guo, Saket Tiwari 외 arxiv

We investigate why deep neural networks suffer from loss of plasticity in continual learning, and thus fail to learn new tasks without reinitializing parameters. We show that this failure is preceded by Hessian spectral …

Reinforcement LearningContinual Learning

SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning

2026-05-06 · Lirui Luo, Guoxi Zhang, Hongming Xu, Cong Fang 외 arxiv

In deep reinforcement learning (DRL), an agent is trained from a stream of experience. In a continual learning setting, such agents can suffer from plasticity loss: their ability to learn new skills from new experiences …

Reinforcement LearningContinual Learning

Reinitializing weights vs units for maintaining plasticity in neural networks

2025-07-31 · J. Fernando Hernandez-Garcia, Shibhansh Dohare, Jun Luo, Rich S. Sutton arxiv

Loss of plasticity is a phenomenon in which a neural network loses its ability to learn when trained for an extended time on non-stationary data. It is a crucial problem to overcome when designing systems that learn cont…

Modulate Your Spectrum in Self-Supervised Learning

2023-05-26 · Xi Weng, Yunhao Ni, Tengwei Song, Jie Luo 외

Whitening loss offers a theoretical guarantee against feature collapse in self-supervised learning (SSL) with joint embedding architectures. Typically, it involves a hard whitening approach, transforming the embedding an…

object-detectionObject DetectionSelf-Supervised Learning

Can We Understand Plasticity Through Neural Collapse?

2024-04-03 · Guglielmo Bonifazi, Iason Chalas, Gian Hess, Jakub Łucki

This paper explores the connection between two recently identified phenomena in deep learning: plasticity loss and neural collapse. We analyze their correlation in different scenarios, revealing a significant association…