Zero-shot learning with partial attributes

MATÍAS MOLINA; JORGE A. SANCHEZ

Merida

Simposio; International Symposium on Intelligent Computing Systems; 2018

Zero-shot classification, i.e. the task of learning predictors forclasses with no training samples, requires to resort to subsidiary infor-mation in order to overcome the lack of annotated data. In the literature,one of the most popular approaches is to represent the class informationby a set of visual attributes and to learn a visual-semantic embeddingthat allow us to transfer the information from those classes with plenty of annotated samples to those with no data available a training time. One mayor limitation of the attribute-based approach is that adding a new class requires a non-negligible annotation effort. This has motivated the search for alternative sources of semantic information. Here, the use of word embeddings learned from raw text appears as an appealing and scalable choice. In this paper, we consider a middleground scenario inwhich attribute vectors are only available for the training categories. We propose a deterministic approach to infer the attributes for the testing classes which, despite its simplicity, shows competitive results. We also propose two simple improvements to the structured embedding formulation of Akata et al., leading to significant improvements on attribute-only classification . Experiments on the Animals With Attributes and Caltech-UCSD Birds datasets show competitive performance.