When BERT Plays the Lottery, All Tickets Are Winning

Sai Prasanna, Anna Rogers, Anna Rumshisky

    Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

    Abstract

    Much of the recent success in NLP is due to the large Transformer-based models such as BERT (Devlin et al, 2019). However, these models have been shown to be reducible to a smaller number of self-attention heads and layers. We consider this phenomenon from the perspective of the lottery ticket hypothesis. For fine-tuned BERT, we show that (a) it is possible to find a subnetwork of elements that achieves performance comparable with that of the full model, and (b) similarly-sized subnetworks sampled from the rest of the model perform worse. However, the "bad" subnetworks can be fine-tuned separately to achieve only slightly worse performance than the "good" ones, indicating that most weights in the pre-trained BERT are potentially useful. We also show that the "good" subnetworks vary considerably across GLUE tasks, opening up the possibilities to learn what knowledge BERT actually uses at inference time.
    Original languageEnglish
    Title of host publicationProceedings of EMNLP
    Number of pages22
    Place of PublicationOnline
    PublisherAssociation for Computational Linguistics
    Publication date1 Nov 2020
    Pages3208-3229
    Publication statusPublished - 1 Nov 2020

    Fingerprint

    Dive into the research topics of 'When BERT Plays the Lottery, All Tickets Are Winning'. Together they form a unique fingerprint.

    Cite this