Preference Learning and Ranking

Preference learning refers to the task of learning to predict (contextualized) preferences on a collection of alternatives, which are often represented in the form of an order relation, on the basis of observed or revealed preference information. Supervision in preference learning is typically weak, in the sense that only partial information about sought structures or indirect information about an underlying value function are provided; a common example is feedback in the form of pairwise comparisons between alternatives. Especially important in preference learning are ranking problems, in which preferences are represented in terms of total or partial order relations. Such problems can be approached in two fundamentally different ways, either by learning binary preferences on pairs of alternatives or by inducing an underlying (latent) value function on single alternatives.

[1]  M. Kendall.  A NEW MEASURE OF RANK CORRELATION , 1938 .

[2]  Peter C. Fishburn,et al.  Utility theory for decision making , 1970 .

[3]  Gerald Tesauro,et al.  Connectionist Learning of Expert Preferences by Comparison Training , 1988, NIPS.

[4]  Yoram Singer,et al.  Learning to Order Things , 1997, NIPS.

[5]  K. Obermayer,et al.  Supervised learning of preference relations , 1998 .

[6]  Thorsten Joachims,et al.  Optimizing search engines using clickthrough data , 2002, KDD.

[7]  Dan Roth,et al.  Constraint Classification: A New Approach to Multiclass Classification , 2002, ALT.

[8]  Editors , 2003 .

[9]  Craig Boutilier,et al.  CP-nets: a tool for represent-ing and reasoning with conditional ceteris paribus state-ments , 2004 .

[10]  Thomas Hofmann,et al.  Large Margin Methods for Structured and Interdependent Output Variables , 2005, J. Mach. Learn. Res..

[11]  Tie-Yan Liu,et al.  Learning to rank: from pairwise approach to listwise approach , 2007, ICML '07.

[12]  Eyke Hüllermeier,et al.  Label ranking by learning pairwise preferences , 2008, Artif. Intell..

[13]  Eyke Hüllermeier,et al.  Binary Decomposition Methods for Multipartite Ranking , 2009, ECML/PKDD.

[14]  Eyke Hüllermeier,et al.  Decision tree and instance-based learning for label ranking , 2009, ICML '09.

[15]  Thomas Gärtner,et al.  Label Ranking Algorithms: A Survey , 2010, Preference Learning.

[16]  Shotaro Akaho,et al.  A Survey and Empirical Comparison of Object Ranking Methods , 2010, Preference Learning.

[17]  Eyke Hüllermeier,et al.  Label Ranking Methods based on the Plackett-Luce Model , 2010, ICML.

[18]  C. Spearman.  The proof and measurement of association between two things. , 2015, International journal of epidemiology.

[19]  Eyke Hüllermeier,et al.  On predictive accuracy and risk minimization in pairwise label ranking , 2010, J. Comput. Syst. Sci..

[20]  Eyke Hüllermeier,et al.  Preference Learning: An Introduction , 2010, Preference Learning.

[21]  Eyke Hüllermeier,et al.  Preferences in AI: An overview , 2011, Artif. Intell..

[22]  Tie-Yan Liu,et al.  Learning to Rank for Information Retrieval , 2011 .

[23]  Eyke Hüllermeier,et al.  Label Ranking with Partial Abstention based on Thresholded Probabilistic Models , 2012, NIPS.

[24]  Eyke Hüllermeier,et al.  A Survey of Preference-Based Online Learning with Bandit Algorithms , 2014, ALT.

[25]  Yangguang Liu,et al.  A Taxonomy of Label Ranking Algorithms , 2014, J. Comput..

[26]  Eyke Hüllermeier,et al.  Dyad Ranking Using a Bilinear Plackett-Luce Model , 2015, LWA.