Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding

no code implementations • 17 Jan 2024 • Depeng Li, Tianqi Wang, Junwei Chen, Qining Ren, Kenji Kawaguchi, Zhigang Zeng

Deep neural networks are susceptible to catastrophic forgetting when trained on sequential tasks.

Continual Learning

Complementary Learning Subnetworks for Parameter-Efficient Class-Incremental Learning

no code implementations • 21 Jun 2023 • Depeng Li, Zhigang Zeng

The inaccessibility of historical data is tackled by holistically controlling the parameters of a well-trained model, ensuring that the decision boundary learned fits new classes while retaining recognition of previously learned classes.

Class Incremental Learning Decision Making +1

IF2Net: Innately Forgetting-Free Networks for Continual Learning

no code implementations • 18 Jun 2023 • Depeng Li, Tianqi Wang, Bingrong Xu, Kenji Kawaguchi, Zhigang Zeng, Ponnuthurai Nagaratnam Suganthan

Continual learning can incrementally absorb new concepts without interfering with previously learned knowledge.

Continual Learning Decision Making

Multi-View Class Incremental Learning

no code implementations • 16 Jun 2023 • Depeng Li, Tianqi Wang, Junwei Chen, Kenji Kawaguchi, Cheng Lian, Zhigang Zeng

Multi-view learning (MVL) has gained great success in integrating information from multiple perspectives of a dataset to improve downstream task performance.

Class Incremental Learning Decision Making +3