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

Authors: Huiping Zhuang, Run He, Kai Tong, Ziqian Zeng, Cen Chen, Zhiping Lin

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 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, Image Net100 and Image Net-Full.
Researcher Affiliation Academia Huiping Zhuang1, Run He1, Kai Tong1, Ziqian Zeng1, Cen Chen2,3*, Zhiping Lin4 1Shien-Ming Wu School of Intelligent Engineering, South China University of Technology, China 2 School of Future Technology, South China University of Technology, China 3 Pazhou Laboratory, China 4 School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore {hpzhuang, zqzeng, chencen}@scut.edu.cn, {wirh, wikaitong}@mail.scut.edu.cn, ezplin@ntu.edu.sg
Pseudocode Yes The DS-AL is summarized in algorithm framework (we place the algorithm in Supplementary material B.
Open Source Code Yes Our codes are available at https: //github.com/ZHUANGHP/Analytic-continual-learning.
Open Datasets Yes Datasets include CIFAR-100, Image Net100 and Image Net-Full.
Dataset Splits No The paper describes training data for each phase (Dtrain k) and testing datasets (Dtest k) but does not explicitly detail a separate validation dataset split.
Hardware Specification No Not found. The paper does not specify the hardware used for the experiments (e.g., GPU/CPU models, memory).
Software Dependencies No Not found in the main text. The paper refers to training strategies from ACIL in supplementary material but does not list specific software dependencies with version numbers.
Experiment Setup Yes Hyperparameters. Two unique hyperparameters (i.e., σC and C) have been introduced in this paper. We utilize grid search to determine their values. ... The activation function chosen in DS-AL is Tanh. The compensation ratios C = 0.6, 0.8, 1.4 on CIFAR-100, Image Net-100 and Image Net-Full respectively.