Attentive Recurrent Comparators

Pranav Shyam, Shubham Gupta, Ambedkar Dukkipati
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:3173-3181, 2017.

Abstract

Rapid learning requires flexible representations to quickly adopt to new evidence. We develop a novel class of models called Attentive Recurrent Comparators (ARCs) that form representations of objects by cycling through them and making observations. Using the representations extracted by ARCs, we develop a way of approximating a dynamic representation space and use it for one-shot learning. In the task of one-shot classification on the Omniglot dataset, we achieve the state of the art performance with an error rate of 1.5\%. This represents the first super-human result achieved for this task with a generic model that uses only pixel information.

Cite this Paper


BibTeX
@InProceedings{pmlr-v70-shyam17a, title = {Attentive Recurrent Comparators}, author = {Pranav Shyam and Shubham Gupta and Ambedkar Dukkipati}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {3173--3181}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/shyam17a/shyam17a.pdf}, url = {https://proceedings.mlr.press/v70/shyam17a.html}, abstract = {Rapid learning requires flexible representations to quickly adopt to new evidence. We develop a novel class of models called Attentive Recurrent Comparators (ARCs) that form representations of objects by cycling through them and making observations. Using the representations extracted by ARCs, we develop a way of approximating a dynamic representation space and use it for one-shot learning. In the task of one-shot classification on the Omniglot dataset, we achieve the state of the art performance with an error rate of 1.5\%. This represents the first super-human result achieved for this task with a generic model that uses only pixel information.} }
Endnote
%0 Conference Paper %T Attentive Recurrent Comparators %A Pranav Shyam %A Shubham Gupta %A Ambedkar Dukkipati %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-shyam17a %I PMLR %P 3173--3181 %U https://proceedings.mlr.press/v70/shyam17a.html %V 70 %X Rapid learning requires flexible representations to quickly adopt to new evidence. We develop a novel class of models called Attentive Recurrent Comparators (ARCs) that form representations of objects by cycling through them and making observations. Using the representations extracted by ARCs, we develop a way of approximating a dynamic representation space and use it for one-shot learning. In the task of one-shot classification on the Omniglot dataset, we achieve the state of the art performance with an error rate of 1.5\%. This represents the first super-human result achieved for this task with a generic model that uses only pixel information.
APA
Shyam, P., Gupta, S. & Dukkipati, A.. (2017). Attentive Recurrent Comparators. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:3173-3181 Available from https://proceedings.mlr.press/v70/shyam17a.html.

Related Material