Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning

About

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.

Huiping Zhuang, Run He, Kai Tong, Ziqian Zeng, Cen Chen, Zhiping Lin• 2024

Related benchmarks

TaskDatasetResultRank
Class-incremental learningCIFAR-100 B0_Inc5
Average Accuracy90.24
63
Class-incremental learningVTAB B0 Inc10
Last Accuracy88.17
54
Class-incremental learningImageNet-100 B=50, C=10 1.0
Avg Incremental Acc75.2
42
Class-incremental learningCIFAR-100 B50Inc10
Avg Accuracy0.6654
41
Incremental LearningCIFAR-100 (test)
Average Accuracy68.4
27
Class-incremental learningImageNet-R B0 Inc20 (test)
Average Performance (A-bar)73.88
23
Cross-Domain Class-Incremental LearningOfficeHome 13 tasks
Accuracy (SD)44.26
18
Cross-Domain Class-Incremental LearningPACS 3 tasks
Accuracy (SD)57.24
18
Cross-Domain Class-Incremental LearningOfficeHome 5 tasks
Average Accuracy (SD)47.92
18
Class-incremental learningDomainNet 5-task
Average Accuracy (SD)32.91
18
Showing 10 of 34 rows

Other info

Follow for update