Decision Boundary-aware Knowledge Consolidation Generates Better Instance-Incremental Learner

TL;DR

This paper introduces a decision boundary-aware knowledge consolidation method for instance-incremental learning, effectively addressing catastrophic forgetting and concept drift without access to old data, and demonstrates its superiority on standard benchmarks.

Contribution

It proposes a novel decision boundary-aware distillation approach for IIL, emphasizing knowledge consolidation from teacher to student, and establishes new benchmarks on Cifar-100 and ImageNet.

Findings

Teacher models outperform student models as incremental learners.

The method effectively mitigates catastrophic forgetting.

Benchmarks on Cifar-100 and ImageNet validate the approach.

Abstract

Instance-incremental learning (IIL) focuses on learning continually with data of the same classes. Compared to class-incremental learning (CIL), the IIL is seldom explored because IIL suffers less from catastrophic forgetting (CF). However, besides retaining knowledge, in real-world deployment scenarios where the class space is always predefined, continual and cost-effective model promotion with the potential unavailability of previous data is a more essential demand. Therefore, we first define a new and more practical IIL setting as promoting the model's performance besides resisting CF with only new observations. Two issues have to be tackled in the new IIL setting: 1) the notorious catastrophic forgetting because of no access to old data, and 2) broadening the existing decision boundary to new observations because of concept drift. To tackle these problems, our key insight is to moderately broaden the decision boundary to fail cases while retain old boundary. Hence, we propose a novel decision boundary-aware distillation method with consolidating knowledge to teacher to ease the student learning new knowledge. We also establish the benchmarks on existing datasets Cifar-100 and ImageNet. Notably, extensive experiments demonstrate that the teacher model can be a better incremental learner than the student model, which overturns previous knowledge distillation-based methods treating student as the main role.