Details
Presenter(s)
![Danilo Vucetic Headshot](https://confcats-catavault.s3.amazonaws.com/CATAVault/ieeecass/master/files/styles/cc_user_photo/s3/user-pictures/21901_0.jpg?h=038a9462&itok=3FXYNp15)
Display Name
Danilo Vucetic
- Affiliation
-
AffiliationMcGill University
- Country
-
CountryCanada
Abstract
We propose a method to reduce the fine-tuning costs of BERT-based natural language models such that they may be realized on resource-constrained devices. We identify memory usage as a major bottleneck and reduce memory operations during fine-tuning by training only a subset of the model. A reconfiguration of the model achieves better memory performance and training time. Our approach reduces memory usage, memory access time, and fine-tuning time substantially, while achieving near-baseline metric performance.