Self-Cooperation Knowledge Distillation for Novel Class Discovery
This work addresses a domain-specific issue in Novel Class Discovery by improving model balance between reviewing known classes and discovering novel ones, though it is incremental as it builds on existing representation space methods.
The paper tackles the problem of imbalanced sample sizes between known and novel classes in Novel Class Discovery, which causes models to favor dominant classes, and proposes a Self-Cooperation Knowledge Distillation method that achieves state-of-the-art performance on six datasets.
Novel Class Discovery (NCD) aims to discover unknown and novel classes in an unlabeled set by leveraging knowledge already learned about known classes. Existing works focus on instance-level or class-level knowledge representation and build a shared representation space to achieve performance improvements. However, a long-neglected issue is the potential imbalanced number of samples from known and novel classes, pushing the model towards dominant classes. Therefore, these methods suffer from a challenging trade-off between reviewing known classes and discovering novel classes. Based on this observation, we propose a Self-Cooperation Knowledge Distillation (SCKD) method to utilize each training sample (whether known or novel, labeled or unlabeled) for both review and discovery. Specifically, the model's feature representations of known and novel classes are used to construct two disjoint representation spaces. Through spatial mutual information, we design a self-cooperation learning to encourage model learning from the two feature representation spaces from itself. Extensive experiments on six datasets demonstrate that our method can achieve significant performance improvements, achieving state-of-the-art performance.