We introduce a new model for online ranking in which the click probability factors into an examination and attractiveness function and the attractiveness function is a linear function of a feature vector and an unknown parameter. Only relatively mild assumptions are made on the examination function. A novel algorithm for this setup is analysed, showing that the dependence on the number of items is replaced by a dependence on the dimension, allowing the new algorithm to handle a large number of items. When reduced to the orthogonal case, the regret of the algorithm improves on the state-of-the-art.
%0 Conference Paper
%1 LLSz19
%A Li, S.
%A Lattimore, T.
%A Szepesvári, Cs.
%B ICML
%D 2019
%K bandits, finite-armed information, learning learning, linear online partial rank, ranking, stochastic to
%P 3856--3865
%T Online Learning to Rank with Features
%X We introduce a new model for online ranking in which the click probability factors into an examination and attractiveness function and the attractiveness function is a linear function of a feature vector and an unknown parameter. Only relatively mild assumptions are made on the examination function. A novel algorithm for this setup is analysed, showing that the dependence on the number of items is replaced by a dependence on the dimension, allowing the new algorithm to handle a large number of items. When reduced to the orthogonal case, the regret of the algorithm improves on the state-of-the-art.
@inproceedings{LLSz19,
abstract = {We introduce a new model for online ranking in which the click probability factors into an examination and attractiveness function and the attractiveness function is a linear function of a feature vector and an unknown parameter. Only relatively mild assumptions are made on the examination function. A novel algorithm for this setup is analysed, showing that the dependence on the number of items is replaced by a dependence on the dimension, allowing the new algorithm to handle a large number of items. When reduced to the orthogonal case, the regret of the algorithm improves on the state-of-the-art.},
added-at = {2020-03-17T03:03:01.000+0100},
author = {Li, S. and Lattimore, T. and {Sz}epesv{\'a}ri, {Cs}.},
bdsk-url-1 = {http://proceedings.mlr.press/v54/hanawal17a.html},
biburl = {https://www.bibsonomy.org/bibtex/293a1024c3d9c9eefc67af48e0b79ff7f/csaba},
booktitle = {ICML},
date-added = {2019-07-20 13:48:35 -0600},
date-modified = {2019-07-20 13:51:47 -0600},
interhash = {e3ff79de4b5b37e3ee4a0f35eb2fcd00},
intrahash = {93a1024c3d9c9eefc67af48e0b79ff7f},
keywords = {bandits, finite-armed information, learning learning, linear online partial rank, ranking, stochastic to},
month = May,
pages = {3856--3865},
pdf = {papers/ICML2019-RecurRank.pdf},
timestamp = {2020-03-17T03:03:01.000+0100},
title = {Online Learning to Rank with Features},
year = 2019
}