Bradley Terry Ranking
Medium · Solved in NumPy · Machine learning coding practice
Problem
Implement a Bradley-Terry rating system to learn competitor ratings from pairwise match results.
The Bradley-Terry model is used to estimate the skill levels of competitors based on the outcomes of head-to-head matches. Given a set of match results, your task is to compute the ratings that best explain the observed outcomes.
Topics: optimization, statistics, gradient-descent, pairwise-comparison, ranking, numpy, pytorch
The full statement, worked examples, hints and the test suite are available once you sign in. You can then solve Bradley Terry Ranking in the browser and run it against the tests.