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

Sparse Orthogonal Parameters Tuning for Continual Learning

About

Continual learning methods based on pre-trained models (PTM) have recently gained attention which adapt to successive downstream tasks without catastrophic forgetting. These methods typically refrain from updating the pre-trained parameters and instead employ additional adapters, prompts, and classifiers. In this paper, we from a novel perspective investigate the benefit of sparse orthogonal parameters for continual learning. We found that merging sparse orthogonality of models learned from multiple streaming tasks has great potential in addressing catastrophic forgetting. Leveraging this insight, we propose a novel yet effective method called SoTU (Sparse Orthogonal Parameters TUning). We hypothesize that the effectiveness of SoTU lies in the transformation of knowledge learned from multiple domains into the fusion of orthogonal delta parameters. Experimental evaluations on diverse CL benchmarks demonstrate the effectiveness of the proposed approach. Notably, SoTU achieves optimal feature representation for streaming data without necessitating complex classifier designs, making it a Plug-and-Play solution.

Kun-Peng Ning, Hai-Jian Ke, Yu-Yang Liu, Jia-Yu Yao, Yong-Hong Tian, Li Yuan• 2024

Related benchmarks

TaskDatasetResultRank
Class-incremental learningCIFAR-100
Average Accuracy94.5
150
Class-incremental learningImageNet-R
Last Accuracy79.5
147
Class-incremental learningImageNet A
Average Accuracy75.1
110
Class-incremental learningCUB200
Last Accuracy89.1
64
Class-incremental learningVTAB
Avg Accuracy96.7
55
Class-incremental learningCARS196
Final Accuracy78.8
15
Showing 6 of 6 rows

Other info

Follow for update