Loss-landscape visualization of BERT fine-tuning indicates pre-training provides a better initialization that yields flatter optima, faster convergence, and lower layers that are more invariant across tasks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Visualizing and Understanding the Effectiveness of BERT
Loss-landscape visualization of BERT fine-tuning indicates pre-training provides a better initialization that yields flatter optima, faster convergence, and lower layers that are more invariant across tasks.