Online Submodular Maximization under a Matroid Constraint with Application to Learning Assignments

Which ads should we display in sponsored search in order to maximize our revenue? How should we dynamically rank information sources to maximize the value of the ranking? These applications exhibit strong diminishing returns: Redundancy decreases the marginal utility of each ad or information source. We show that these and other problems can be formalized as repeatedly selecting an assignment of items to positions to maximize a sequence of monotone submodular functions that arrive one by one. We present an efficient algorithm for this general problem and analyze it in the no-regret model. Our algorithm possesses strong theoretical guarantees, such as a performance ratio that converges to the optimal constant of 1 1=e. We empirically evaluate our algorithm on two real-world online optimization problems on the web: ad allocation with submodular utilities, and dynamically ranking blogs to detect information cascades. Finally, we present a second algorithm that handles the more general case in which the feasible sets are given by a matroid constraint, while still maintaining a 1 1=e asymptotic performance ratio.

[1]  M. L. Fisher,et al.  An analysis of approximations for maximizing submodular set functions—I , 1978, Math. Program..

[2]  Jan Vondrák,et al.  Optimal approximation for the submodular welfare problem in the value oracle model , 2008, STOC.

[3]  Nicholas Roy,et al.  Contextual Sequence Prediction with Application to Control Library Optimization , 2013 .

[4]  Andreas Krause,et al.  Cost-effective outbreak detection in networks , 2007, KDD '07.

[5]  Adam Tauman Kalai,et al.  Playing Games with Approximation Algorithms , 2009, SIAM J. Comput..

[6]  Filip Radlinski,et al.  Learning diverse rankings with multi-armed bandits , 2008, ICML '08.

[7]  Jon Feldman,et al.  Sponsored Search Auctions with Markovian Users , 2008, WINE.

[8]  Vahab S. Mirrokni,et al.  Tight information-theoretic lower bounds for welfare maximization in combinatorial auctions , 2008, EC '08.

[9]  Andreas Krause,et al.  Submodular Function Maximization , 2014, Tractability.

[10]  Jan Vondrák,et al.  Submodularity in Combinatorial Optimization , 2007 .

[11]  Yisong Yue,et al.  Learning Policies for Contextual Submodular Prediction , 2013, ICML.

[12]  Jan Vondrák,et al.  Maximizing a Monotone Submodular Function Subject to a Matroid Constraint , 2011, SIAM J. Comput..

[13]  Alexander Schrijver,et al.  Combinatorial optimization. Polyhedra and efficiency. , 2003 .

[14]  Yishay Mansour,et al.  From External to Internal Regret , 2005, J. Mach. Learn. Res..

[15]  Peter Auer,et al.  The Nonstochastic Multiarmed Bandit Problem , 2002, SIAM J. Comput..

[16]  Andreas Krause,et al.  Online Learning of Assignments that Maximize Submodular Functions , 2009, ArXiv.

[17]  Zhen Liu,et al.  Performance Modeling and Engineering , 2010 .

[18]  Jon Feldman,et al.  Algorithmic Methods for Sponsored Search Advertising , 2008, ArXiv.

[19]  Andreas Krause,et al.  Online Learning of Assignments , 2009, NIPS.

[20]  R. Vohra,et al.  Algorithmic Game Theory: Sponsored Search Auctions , 2007 .

[21]  Jan Vondrák,et al.  Maximizing a Submodular Set Function Subject to a Matroid Constraint (Extended Abstract) , 2007, IPCO.

[22]  Matthew J. Streeter,et al.  An Online Algorithm for Maximizing Submodular Functions , 2008, NIPS.

[23]  Santosh S. Vempala,et al.  Efficient algorithms for online decision problems , 2005, Journal of computer and system sciences (Print).

[24]  Andreas Krause,et al.  Submodularity and its applications in optimized information gathering , 2011, TIST.