Differentiable Subset Pruning of Transformer Heads.

Jiaoda Li Ryan Cotterell Mrinmaya Sachan

Published in: CoRR (2021)

Keyphrases

fuzzy logic
search space
fault diagnosis
information systems
image sequences
objective function
tree pruning
pruning method
pruning methods
magnetic recording
high voltage
power system
incipient fault
pruning algorithms
power transformers
evolutionary algorithm
search algorithm
data structure
decision trees