First direct comparison finds counterfactual LTR best with low bias or noise but potentially harmful otherwise, while online LTR is robust yet requires ranking control.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
To Model or to Intervene: A Comparison of Counterfactual and Online Learning to Rank from User Interactions
First direct comparison finds counterfactual LTR best with low bias or noise but potentially harmful otherwise, while online LTR is robust yet requires ranking control.