用物理熔化模型解决持续学习中的记忆遗忘问题
Continual Learning as a Multiphase Moving-Boundary Problem
- 将已学知识视为固态，未用容量为液态，边界随学习扩展
- 冻结内部知识使遗忘率接近零，性能媲美存储原始数据的模型
- 适合研究持续学习、神经网络记忆机制的学者参考
持续学习难以平衡保留旧知识与吸收新任务之间的矛盾。Stefan-CL 通过熔化物理原理优雅地解决稳定性-可塑性困境：将已巩固的知识视为受保护的‘固态’，未使用的容量则视为可调节的‘液态’。随着网络学习，这一相变边界逐渐扩展，由一个‘潜热’调节旋钮控制。通过数学上冻结已学内部区域，Stefan-CL 将遗忘降至近零，其性能可媲美需要存储原始数据的记忆密集型基线模型，为人工智能提供了一条基于物理规律的优美持续学习路径。
原文摘要 · Abstract (English)
Continual learning struggles to balance retaining past knowledge with absorbing new tasks. Stefan-CL elegantly resolves this stability-plasticity dilemma through the physics of melting. It frames consolidated knowledge as a protected "solid" and unused capacity as an adaptable "liquid." As the network learns, this boundary expands, governed by a "latent heat" tuning dial. By mathematically freezing the learned interior, Stefan-CL cuts forgetting to near zero, matching memory-heavy baselines without storing raw data, forging a beautiful, physics-grounded path for AI.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。