Optimisation methods for ranking functions with multiple parameters

  • Authors:
  • Michael Taylor;Hugo Zaragoza;Nick Craswell;Stephen Robertson;Chris Burges

  • Affiliations:
  • Microsoft Research, Cambridge, UK;Yahoo! Research, Barcelona, Spain;Microsoft Research, Cambridge, UK;Microsoft Research, Cambridge, UK;Microsoft Research, Redmond, WA

  • Venue:
  • CIKM '06 Proceedings of the 15th ACM international conference on Information and knowledge management
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Optimising the parameters of ranking functions with respect to standard IR rank-dependent cost functions has eluded satisfactory analytical treatment. We build on recent advances in alternative differentiable pairwise cost functions, and show that these techniques can be successfully applied to tuning the parameters of an existing family of IR scoring functions (BM25), in the sense that we cannot do better using sensible search heuristics that directly optimize the rank-based cost function NDCG. We also demonstrate how the size of training set affects the number of parameters we can hope to tune this way.