通过三阶段巩固进行持续学习
Continual Learning by Three-Phase Consolidation
AI总结：
本文提出TPC（三阶段巩固）方法，通过三个不同学习阶段消除类偏差并限制梯度修正，有效控制持续学习中的遗忘问题，在复杂数据集上展现出优于现有方法的准确性与效率。
摘要：
本文提出TPC（Three-Phase Consolidation）作为一种简单但有效的方法，用于持续学习新类（和/或已知类的实例），同时控制对先前知识的遗忘。每个经验（又称任务）在三个阶段中学习，这些阶段以不同的规则和学习动态为特征，旨在消除类偏差问题（由于类不平衡引起），并限制基于梯度的修正以防止对代表性不足的类的遗忘。在复杂数据集上的多项实验表明，与竞争性现有方法相比，该方法在准确性和效率上具有优势。该算法和本文提出的所有结果均可完全复现，这得益于其在持续学习的Avalanche开放框架上的发布。
英文摘要：
TPC (Three-Phase Consolidation) is here introduced as a simple but effective approach to continually learn new classes (and/or instances of known classes) while controlling forgetting of previous knowledge. Each experience (a.k.a. task) is learned in three phases characterized by different rules and learning dynamics, aimed at removing the class-bias problem (due to class unbalancing) and limiting gradient-based corrections to prevent forgetting of underrepresented classes. Several experiments on complex datasets demonstrate its accuracy and efficiency advantages over competitive existing approaches. The algorithm and all the results presented in this paper are fully reproducible thanks to its publication on the Avalanche open framework for continual learning.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
