title = "Learning Optimal Auction Mechanism in Sponsored
Search",
booktitle = "Twenty-third International Conference on Artificial
Intelligence, IJCAI 2013",
year = "2013",
editor = "Francesca Rossi",
address = "Beijing, China",
month = aug # " 3-9",
keywords = "genetic algorithms, genetic programming, computer
science - computer science and game theory, computer
science - learning",
URL = "http://arxiv.org/abs/1406.0728",
abstract = "Sponsored search is an important monetization channel
for search engines, in which an auction mechanism is
used to select the ads shown to users and determine the
prices charged from advertisers. There have been
several pieces of work in the literature that
investigate how to design an auction mechanism in order
to optimise the revenue of the search engine. However,
due to some unrealistic assumptions used, the practical
values of these studies are not very clear. In this
paper, we propose a novel game-theoretic machine
learning approach, which naturally combines machine
learning and game theory, and learns the auction
mechanism using a bilevel optimisation framework. In
particular, we first learn a Markov model from
historical data to describe how advertisers change
their bids in response to an auction mechanism, and
then for any given auction mechanism, we use the learnt
model to predict its corresponding future bid
sequences. Next we learn the auction mechanism through
empirical revenue maximisation on the predicted bid
sequences. We show that the empirical revenue will
converge when the prediction period approaches
infinity, and a Genetic Programming algorithm can
effectively optimise this empirical revenue. Our
experiments indicate that the proposed approach is able
to produce a much more effective auction mechanism than
several baselines.
abstract from oai:arXiv.org:1406.0728",
notes = "see also A Game-theoretic Machine Learning Approach
for Revenue Maximization in Sponsored Search
\cite{oai:arXiv.org:1406.0728}
http://ijcai13.org/program/accepted_papers",