We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CL

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Pruning Pretrained Encoders with a Multitask Objective

Abstract: The sizes of pretrained language models make them challenging and expensive to use when there are multiple desired downstream tasks. In this work, we adopt recent strategies for model pruning during finetuning to explore the question of whether it is possible to prune a single encoder so that it can be used for multiple tasks. We allocate a fixed parameter budget and compare pruning a single model with a multitask objective against the best ensemble of single-task models. We find that under two pruning strategies (element-wise and rank pruning), the approach with the multitask objective outperforms training models separately when averaged across all tasks, and it is competitive on each individual one. Additional analysis finds that using a multitask objective during pruning can also be an effective method for reducing model sizes for low-resource tasks.
Comments: ENLSP NeurIPS 2021
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Cite as: arXiv:2112.05705 [cs.CL]
  (or arXiv:2112.05705v1 [cs.CL] for this version)

Submission history

From: Patrick Xia [view email]
[v1] Fri, 10 Dec 2021 17:57:33 GMT (286kb,D)

Link back to: arXiv, form interface, contact.