Demystifying the Base and Novel Performances for Few-shot Class-incremental Learning
This work addresses incremental learning scenarios where new classes with few samples are added, but it is incremental as it builds on existing methods without major breakthroughs.
The paper tackles the problem of few-shot class-incremental learning by analyzing how base and novel performances vary with different model parameters, and proposes a simple method called NoNPC that uses normalized prototype classifiers without training for novel classes, achieving comparable performance to state-of-the-art algorithms.
Few-shot class-incremental learning (FSCIL) has addressed challenging real-world scenarios where unseen novel classes continually arrive with few samples. In these scenarios, it is required to develop a model that recognizes the novel classes without forgetting prior knowledge. In other words, FSCIL aims to maintain the base performance and improve the novel performance simultaneously. However, there is little study to investigate the two performances separately. In this paper, we first decompose the entire model into four types of parameters and demonstrate that the tendency of the two performances varies greatly with the updated parameters when the novel classes appear. Based on the analysis, we propose a simple method for FSCIL, coined as NoNPC, which uses normalized prototype classifiers without further training for incremental novel classes. It is shown that our straightforward method has comparable performance with the sophisticated state-of-the-art algorithms.