Locality and Compositionality in Zero-Shot Learning

Tristan Sylvain, Linda Petrini, Devon Hjelm

TL;DR: An analysis of the effects of compositionality and locality on representation learning for zero-shot learning.

Abstract:

In this work we study locality and compositionality in the context of learning representations for Zero Shot Learning (ZSL).
In order to well-isolate the importance of these properties in learned representations, we impose the additional constraint that, differently from most recent work in ZSL, no pre-training on different datasets (e.g. ImageNet) is performed.
The results of our experiment show how locality, in terms of small parts of the input, and compositionality, i.e. how well can the learned representations be expressed as a function of a smaller vocabulary, are both deeply related to generalization and motivate the focus on more local-aware models in future research directions for representation learning.

Keywords: Zero-shot learning, Compositionality, Locality, Deep Learning

Data: [AwA2](https://paperswithcode.com/dataset/awa2-1), [CUB-200-2011](https://paperswithcode.com/dataset/cub-200-2011)