Universality of Mallows' and degeneracy of Kendall's kernels for rankings

  title={Universality of Mallows' and degeneracy of Kendall's kernels for rankings},
  author={Horia Mania and Aaditya Ramdas and Martin J. Wainwright and Michael I. Jordan and Benjamin Recht},
Kernel methods provide an attractive framework for aggrega tin and learning from ranking data, and so understanding the fundamental properties o f kernels over permutations is a question of broad interest. We provide a detailed analysis o f the Fourier spectra of the standard Kendall and Mallows kernels, and a new class of polynomi al-type kernels. We prove that the Kendall kernel has exactly two irreducible representat ions at which the Fourier transform is non-zero, and moreover, the… CONTINUE READING
This paper has been referenced on Twitter 9 times. VIEW TWEETS


Publications referenced by this paper.

and F

  • S. T. Rachev, L. Klebanov, S. V. Stoyanov
  • Fabozzi . The Method of Distances in the Theory…
  • 2013
1 Excerpt

ourier theoretic probabilistic inference over permutations

  • Jonathan Huang, Carlos Guestrin, F LeonidasGuibas.
  • Journal of Machine Learning Research ,
  • 2009
3 Excerpts

Group theoretical methods in machine learning

  • Imre Risi Kondor
  • Columbia University,
  • 2008
2 Excerpts

Individual choice behavior: A theoretical analysis

  • R Duncan Luce
  • Courier Corporation
  • 2005
2 Excerpts

Similar Papers

Loading similar papers…