Novelty-Prepared Few-Shot Classification
This work addresses the data scarceness issue in few-shot classification for real-world applications, offering an incremental improvement by optimizing model capacity for better adaptation to new classes.
The paper tackles the problem of few-shot classification by addressing how pre-trained feature extractors can leave insufficient embedding space for unseen classes, proposing a self-compacting softmax loss (SSL) that prevents full occupancy and leads to significant improvement in state-of-the-art performance on CUB-200-2011 and mini-ImageNet datasets.
Few-shot classification algorithms can alleviate the data scarceness issue, which is vital in many real-world problems, by adopting models pre-trained from abundant data in other domains. However, the pre-training process was commonly unaware of the future adaptation to other concept classes. We disclose that a classically fully trained feature extractor can leave little embedding space for unseen classes, which keeps the model from well-fitting the new classes. In this work, we propose to use a novelty-prepared loss function, called self-compacting softmax loss (SSL), for few-shot classification. The SSL can prevent the full occupancy of the embedding space. Thus the model is more prepared to learn new classes. In experiments on CUB-200-2011 and mini-ImageNet datasets, we show that SSL leads to significant improvement of the state-of-the-art performance. This work may shed some light on considering the model capacity for few-shot classification tasks.