Exploring the Impact of Parameter Update Magnitude on Forgetting and Generalization of Continual Learning
The magnitude of parameter updates are considered a key factor in continual learning. However, most existing studies focus on designing diverse update strategies, while a theoretical understanding of the underlying mechanisms remains limited. Therefore, we characterize model's forgetting from the perspective of parameter update magnitude and formalize it as knowledge degradation induced by task-specific drift in the parameter space, which has not been fully captured in previous studies due to their assumption of a unified parameter space. By deriving the optimal parameter update magnitude that minimizes forgetting, we unify two representative update paradigms, frozen training and initialized training, within an optimization framework for constrained parameter updates. Our theoretical results further reveals that sequence tasks with small parameter distances exhibit better generalization and less forgetting under frozen training rather than initialized training. These theoretical insights inspire a novel hybrid parameter update strategy that adaptively adjusts update magnitude based on gradient directions. Experiments on deep neural networks demonstrate that this hybrid approach outperforms standard training strategies, providing new theoretical perspectives and practical inspiration for designing efficient and scalable continual learning algorithms.
Code (0)
등록된 구현이 없습니다.
Tasks
Continual LearningSimilar Papers 제목 키워드 기반
Mechanism-Guided Selective Unlearning for RLVR-Induced Reasoning
We propose MAST (Mechanism-Aligned Selective Targeting), a mechanism-guided method for unlearning RLVR-induced reasoning with substantially lower collateral damage than standard full-parameter updates. In matched SFT/RLV…
FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning
Continual learning (CL) for large language models (LLMs) aims to enable sequential knowledge acquisition without catastrophic forgetting. Memory replay methods are widely used for their practicality and effectiveness, bu…
Continual LearningMeta Continual Learning
Using neural networks in practical settings would benefit from the ability of the networks to learn new tasks throughout their lifetimes without forgetting the previous tasks. This ability is limited in the current deep …
Continual LearningScaling Laws for Forgetting When Fine-Tuning Large Language Models
We study and quantify the problem of forgetting when fine-tuning pre-trained large language models (LLMs) on a downstream task. We find that parameter-efficient fine-tuning (PEFT) strategies, such as Low-Rank Adapters (L…
parameter-efficient fine-tuningPAT: Pixel-wise Adaptive Training for Long-tailed Segmentation
Beyond class frequency, we recognize the impact of class-wise relationships among various class-specific predictions and the imbalance in label masks on long-tailed segmentation learning. To address these challenges, we …