BriefGPT.xyz
Aug, 2024
通过任务启发式建模的自适应变分持续学习
Adaptive Variational Continual Learning via Task-Heuristic Modelling
HTML
PDF
Fan Yang
TL;DR
本研究解决了变分持续学习(VCL)中固定超参数导致的性能限制问题。提出了一种名为AutoVCL的新方法,通过结合任务启发式信息来自动调整超参数,从而提升模型的学习效率和优化能力。结果表明,相较于标准的GVCL模型,AutoVCL在处理新的任务时表现更佳,具有显著的潜在影响力。
Abstract
Variational
Continual Learning
(VCL) is a turn-key learning algorithm that has state-of-the-art performance among the best
Continual Learning
models. In our work, we explore an extension of the generalized variat
→