Actively Learning Hemimetrics with Applications to Eliciting User Preferences

Motivated by an application of eliciting users' preferences, we investigate the problem of learning hemimetrics, i.e., pairwise distances among a set of $n$ items that satisfy triangle inequalities and non-negativity constraints. In our application, the (asymmetric) distances quantify private costs a user incurs when substituting one item by another. We aim to learn these distances (costs) by asking the users whether they are willing to switch from one item to another for a given incentive offer. Without exploiting structural constraints of the hemimetric polytope, learning the distances between each pair of items requires $\Theta(n^2)$ queries. We propose an active learning algorithm that substantially reduces this sample complexity by exploiting the structural constraints on the version space of hemimetrics. Our proposed algorithm achieves provably-optimal sample complexity for various instances of the task. For example, when the items are embedded into $K$ tight clusters, the sample complexity of our algorithm reduces to $O(n K)$. Extensive experiments on a restaurant recommendation data set support the conclusions of our theoretical analysis.

[1]  Michael C. Hout,et al.  Multidimensional Scaling , 2003, Encyclopedic Dictionary of Archaeology.

[2]  Inderjit S. Dhillon,et al.  The Metric Nearness Problem , 2008, SIAM J. Matrix Anal. Appl..

[3]  M. Vázquez-Gallo,et al.  Active Learning and Dynamic Pricing Policies , 2014 .

[4]  R. Nowak,et al.  Upper and Lower Error Bounds for Active Learning , 2006 .

[5]  Rong Jin,et al.  Distance Metric Learning: A Comprehensive Survey , 2006 .

[6]  Fei Sha,et al.  Similarity Learning for High-Dimensional Sparse Data , 2014, AISTATS.

[7]  David Cohn,et al.  Active Learning , 2010, Encyclopedia of Machine Learning.

[8]  Gert R. G. Lanckriet,et al.  Efficient Learning of Mahalanobis Metrics for Ranking , 2014, ICML.

[9]  Andreas Krause,et al.  Incentivizing Users for Balancing Bike Sharing Systems , 2015, AAAI.

[10]  Matthew Gentzkow,et al.  Competition in Persuasion , 2011 .

[11]  Eric Eaton,et al.  Learning user preferences for sets of objects , 2006, ICML.

[12]  Charles Elkan,et al.  Using the Triangle Inequality to Accelerate k-Means , 2003, ICML.

[13]  Robert D. Nowak,et al.  Low-dimensional embedding using adaptively selected ordinal data , 2011, 2011 49th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[14]  Yishay Mansour,et al.  Learning What's Going on: Reconstructing Preferences and Priorities from Opaque Transactions , 2015, EC.

[15]  Ramesh Johari,et al.  At What Quality and What Price?: Eliciting Buyer Preferences as a Market Design Problem , 2015, EC.

[16]  Michael I. Jordan,et al.  Distance Metric Learning with Application to Clustering with Side-Information , 2002, NIPS.

[17]  Andreas Krause,et al.  Truthful incentives in crowdsourcing tasks using regret minimization mechanisms , 2013, WWW.

[18]  Emir Kamenica,et al.  Bayesian Persuasion , 2009 .

[19]  Eric O. Postma,et al.  Dimensionality Reduction: A Comparative Review , 2008 .

[20]  Matti Kääriäinen,et al.  Active Learning in the Non-realizable Case , 2006, ALT.

[21]  Jingrui He,et al.  Manifold-ranking based image retrieval , 2004, MULTIMEDIA '04.

[22]  Shiguang Shan,et al.  Log-Euclidean Metric Learning on Symmetric Positive Definite Manifold with Application to Image Set Classification , 2015, ICML.

[23]  Robert D. Nowak,et al.  Active Ranking using Pairwise Comparisons , 2011, NIPS.

[24]  Marc Sebban,et al.  A Survey on Metric Learning for Feature Vectors and Structured Data , 2013, ArXiv.

[25]  Chien-Ju Ho,et al.  Low-Cost Learning via Active Data Procurement , 2015, EC.

[26]  Shimon Ullman,et al.  Learning Local Invariant Mahalanobis Distances , 2015, ICML.

[27]  Ehsan Amid,et al.  Multiview Triplet Embedding: Learning Attributes in Multiple Maps , 2015, ICML.

[28]  Adam Tauman Kalai,et al.  Adaptively Learning the Crowd Kernel , 2011, ICML.

[29]  Dan Roth,et al.  Margin-Based Active Learning for Structured Output Spaces , 2006, ECML.