Pessimistic bilevel optimization, which tunes hyperparameters against the worst-case inner-level model, gives more robust binary classifiers than optimistic bilevel tuning under limited or perturbed data.
Weak linear bilevel programming problems: existence of solutions via a penalty method
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Hyperparameter Tuning Through Pessimistic Bilevel Optimization
Pessimistic bilevel optimization, which tunes hyperparameters against the worst-case inner-level model, gives more robust binary classifiers than optimistic bilevel tuning under limited or perturbed data.