Model inference for ranking from pairwise comparisons
This work addresses ranking challenges in domains like sports analytics, though it appears incremental as it builds on standard approaches without introducing a new paradigm.
The paper tackles the problem of ranking objects from noisy pairwise comparisons, such as ranking tennis players from match outcomes, by developing an efficient algorithm that simultaneously infers unobserved strengths and the mapping function from strengths to probabilities, with experimental evidence showing robustness across different model specifications.
We consider the problem of ranking objects from noisy pairwise comparisons, for example, ranking tennis players from the outcomes of matches. We follow a standard approach to this problem and assume that each object has an unobserved strength and that the outcome of each comparison depends probabilistically on the strengths of the comparands. However, we do not assume to know a priori how skills affect outcomes. Instead, we present an efficient algorithm for simultaneously inferring both the unobserved strengths and the function that maps strengths to probabilities. Despite this problem being under-constrained, we present experimental evidence that the conclusions of our Bayesian approach are robust to different model specifications. We include several case studies to exemplify the method on real-world data sets.