Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors.

Ido Amos Jonathan Berant Ankit Gupta

Published in: CoRR (2023)

Keyphrases

data driven
prior knowledge
probabilistic model
bayesian framework
artificial intelligence
information systems
information retrieval
learning algorithm
feature selection
website
case study
database systems
expert systems
model selection
regression model
experimental data