English

AlignNet: Unsupervised Entity Alignment

Computer Vision and Pattern Recognition 2020-07-22 v2 Artificial Intelligence Machine Learning

Abstract

Recently developed deep learning models are able to learn to segment scenes into component objects without supervision. This opens many new and exciting avenues of research, allowing agents to take objects (or entities) as inputs, rather that pixels. Unfortunately, while these models provide excellent segmentation of a single frame, they do not keep track of how objects segmented at one time-step correspond (or align) to those at a later time-step. The alignment (or correspondence) problem has impeded progress towards using object representations in downstream tasks. In this paper we take steps towards solving the alignment problem, presenting the AlignNet, an unsupervised alignment module.

Keywords

Cite

@article{arxiv.2007.08973,
  title  = {AlignNet: Unsupervised Entity Alignment},
  author = {Antonia Creswell and Kyriacos Nikiforou and Oriol Vinyals and Andre Saraiva and Rishabh Kabra and Loic Matthey and Chris Burgess and Malcolm Reynolds and Richard Tanburn and Marta Garnelo and Murray Shanahan},
  journal= {arXiv preprint arXiv:2007.08973},
  year   = {2020}
}