ACL 2021long30 citations

Reservoir Transformers

Sheng Shen, Alexei Baevski, Ari Morcos, Kurt Keutzer, Michael Auli, Douwe Kiela

Abstract

We demonstrate that transformers obtain impressive performance even when some of the layers are randomly initialized and never updated. Inspired by old and well-established ideas in machine learning, we explore a variety of non-linear “reservoir” layers interspersed with regular transformer layers, and show improvements in wall-clock compute time until convergence, as well as overall performance, on various machine translation and (masked) language modelling tasks.

BibTeX
@inproceedings{shen-etal-2021-reservoir,
    title = "Reservoir Transformers",
    author = "Shen, Sheng  and
      Baevski, Alexei  and
      Morcos, Ari  and
      Keutzer, Kurt  and
      Auli, Michael  and
      Kiela, Douwe",
    editor = "Zong, Chengqing  and
      Xia, Fei  and
      Li, Wenjie  and
      Navigli, Roberto",
    booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.acl-long.331/",
    doi = "10.18653/v1/2021.acl-long.331",
    pages = "4294--4309"
}
Reservoir Transformers · ACL 2021