Leveraging Systematic Knowledge of 2D Transformations

arXiv:2206.00893v2 Announce Type: replace-cross
Abstract: The existing deep learning models suffer from out-of-distribution (o.o.d.) performance drop in computer vision tasks. In comparison, humans have a remarkable ability to interpret images, even if the scenes in the images are rare, thanks to the systematicity of acquired knowledge. This work focuses on 1) the acquisition of systematic knowledge of 2D transformations, and 2) architectural components that can leverage the learned knowledge in image classification tasks in an o.o.d. setting. With a new …

abstract acquired acquisition arxiv comparison computer computer vision cs.cv cs.lg deep learning distribution humans images knowledge performance tasks type vision work