Learning Compositional Representations for Few-Shot Recognition
This work addresses the data inefficiency of deep learning for few-shot recognition, offering a step towards bridging the gap with human learning, though it is incremental as it builds on existing regularization and attribute-based approaches.
The paper tackles the problem of few-shot recognition by introducing a regularization technique that decomposes learned representations into compositional parts, using attribute annotations to disentangle feature spaces. The method demonstrates improved learning efficiency, requiring fewer examples to classify novel categories on datasets like CUB-200-2011, SUN397, and ImageNet.
One of the key limitations of modern deep learning approaches lies in the amount of data required to train them. Humans, by contrast, can learn to recognize novel categories from just a few examples. Instrumental to this rapid learning ability is the compositional structure of concept representations in the human brain --- something that deep learning models are lacking. In this work, we make a step towards bridging this gap between human and machine learning by introducing a simple regularization technique that allows the learned representation to be decomposable into parts. Our method uses category-level attribute annotations to disentangle the feature space of a network into subspaces corresponding to the attributes. These attributes can be either purely visual, like object parts, or more abstract, like openness and symmetry. We demonstrate the value of compositional representations on three datasets: CUB-200-2011, SUN397, and ImageNet, and show that they require fewer examples to learn classifiers for novel categories.