Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning
This addresses the issue of underutilizing label information in few-shot learning, which is incremental but enhances classification accuracy for tasks like image recognition.
The paper tackles the problem of simplistic class modeling in few-shot learning by proposing an Absolute-relative Learning paradigm that simultaneously learns similarity and class concepts, improving state-of-the-art models on public datasets with concrete performance gains.
The majority of existing few-shot learning methods describe image relations with binary labels. However, such binary relations are insufficient to teach the network complicated real-world relations, due to the lack of decision smoothness. Furthermore, current few-shot learning models capture only the similarity via relation labels, but they are not exposed to class concepts associated with objects, which is likely detrimental to the classification performance due to underutilization of the available class labels. To paraphrase, children learn the concept of tiger from a few of actual examples as well as from comparisons of tiger to other animals. Thus, we hypothesize that in fact both similarity and class concept learning must be occurring simultaneously. With these observations at hand, we study the fundamental problem of simplistic class modeling in current few-shot learning methods. We rethink the relations between class concepts, and propose a novel Absolute-relative Learning paradigm to fully take advantage of label information to refine the image representations and correct the relation understanding in both supervised and unsupervised scenarios. Our proposed paradigm improves the performance of several the state-of-the-art models on publicly available datasets.