DS-AL：一种用于无样本类增量学习的双流解析学习方法
DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning
AI总结：
本文提出双流解析学习方法DS-AL解决无样本类增量学习问题，通过主干流提供闭式线性解并利用补偿流改善欠拟合，在多数据集上性能媲美基于回放的方法，且具备阶段不变性。
摘要：
在无样本约束下的类增量学习（CIL）提出了重大挑战。遵循此约束的现有方法容易发生灾难性遗忘，且程度远超仍能访问过去样本的基于回放的技术。在本文中，为解决无样本CIL问题，我们提出了一种双流解析学习（DS-AL）方法。DS-AL包含一个提供解析（即闭式）线性解的主干流，以及一个补偿流，后者用于改善因采用线性映射而固有的欠拟合局限。主干流将CIL问题重新定义为级联递归最小二乘（C-RLS）任务，使CIL与其联合学习对应项之间实现等价。补偿流由双激活补偿（DAC）模块控制。该模块使用与主干流不同的激活函数重新激活嵌入，并通过将嵌入投影到主干流线性映射的零空间来寻求拟合补偿。实验结果表明，DS-AL尽管是一种无样本技术，但在包括CIFAR-100、ImageNet-100和ImageNet-Full在内的各种数据集上，其性能可与基于回放的方法相当或更优。此外，C-RLS的等价特性使DS-AL能够以阶段不变的方式执行CIL。一项前所未有的500阶段CIL ImageNet任务证实了这一点，其表现水平与5阶段任务完全相同。代码可在https://github.com/ZHUANGHP/Analytic-continual-learning获取。
英文摘要：
Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic forgetting, far more so than replay-based techniques that retain access to past samples. In this paper, to solve the exemplar-free CIL problem, we propose a Dual-Stream Analytic Learning (DS-AL) approach. The DS-AL contains a main stream offering an analytical (i.e., closed-form) linear solution, and a compensation stream improving the inherent under-fitting limitation due to adopting linear mapping. The main stream redefines the CIL problem into a Concatenated Recursive Least Squares (C-RLS) task, allowing an equivalence between the CIL and its joint-learning counterpart. The compensation stream is governed by a Dual-Activation Compensation (DAC) module. This module re-activates the embedding with a different activation function from the main stream one, and seeks fitting compensation by projecting the embedding to the null space of the main stream's linear mapping. Empirical results demonstrate that the DS-AL, despite being an exemplar-free technique, delivers performance comparable with or better than that of replay-based methods across various datasets, including CIFAR-100, ImageNet-100 and ImageNet-Full. Additionally, the C-RLS' equivalent property allows the DS-AL to execute CIL in a phase-invariant manner. This is evidenced by a never-before-seen 500-phase CIL ImageNet task, which performs on a level identical to a 5-phase one. Our codes are available at https://github.com/ZHUANGHP/Analytic-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。
