End-to-end training of Recurrent Neural Networks (RNNs) have been successfully applied to numerous problems that require processing sequences, such as image captioning, machine translation, and text recognition. However, RNNs often struggle to generalise to sequences longer than the ones encountered during training. In this work, we propose to optimize neural networks explicitly for induction. The idea is to first decompose the problem in a sequence of inductive steps and then to explicitly t...Expand abstract
- Publication status:
- Peer review status:
- Peer reviewed
- Publisher's Version
- British Machine Vision Conference Publisher's website
- Publication date:
- Acceptance date:
- Pubs id:
- Local pid:
- Copyright holder:
- Gupta et al.
- Copyright date:
© 2018. The copyright of this document resides with its authors.
It may be distributed unchanged freely in print or electronic forms. This conference item was presented at the 29th British Machine Vision Conference.
Inductive visual localisation: Factorised training for superior generalisation
Views and Downloads
If you are the owner of this record, you can report an update to it here: Report update to this record