An Active Learning Algorithm for Ranking from Pairwise Preferences with an Almost Optimal Query Complexity

TL;DR

This paper introduces an active learning algorithm for ranking from pairwise preferences that nearly matches the theoretical lower bound on query complexity, effectively addressing a long-standing open problem in the field.

Contribution

The authors develop a provably correct, query-efficient active learning algorithm for ranking from pairwise preferences, improving over existing bounds and adapting recent combinatorial decomposition techniques.

Findings

The algorithm significantly reduces the number of preference queries needed.

It nearly achieves the information-theoretic lower bound on query complexity.

The method can be integrated with SVM relaxations for practical applications.

Abstract

We study the problem of learning to rank from pairwise preferences, and solve a long-standing open problem that has led to development of many heuristics but no provable results for our particular problem. Given a set $V$ of $n$ elements, we wish to linearly order them given pairwise preference labels. A pairwise preference label is obtained as a response, typically from a human, to the question "which if preferred, u or v?$ for two elements $u,v\in V$. We assume possible non-transitivity paradoxes which may arise naturally due to human mistakes or irrationality. The goal is to linearly order the elements from the most preferred to the least preferred, while disagreeing with as few pairwise preference labels as possible. Our performance is measured by two parameters: The loss and the query complexity (number of pairwise preference labels we obtain). This is a typical learning problem, with the exception that the space from which the pairwise preferences is drawn is finite, consisting of ${n\choose 2}$ possibilities only. We present an active learning algorithm for this problem, with query bounds significantly beating general (non active) bounds for the same error guarantee, while almost achieving the information theoretical lower bound. Our main construct is a decomposition of the input s.t. (i) each block incurs high loss at optimum, and (ii) the optimal solution respecting the decomposition is not much worse than the true opt. The decomposition is done by adapting a recent result by Kenyon and Schudy for a related combinatorial optimization problem to the query efficient setting. We thus settle an open problem posed by learning-to-rank theoreticians and practitioners: What is a provably correct way to sample preference labels? To further show the power and practicality of our solution, we show how to use it in concert with an SVM relaxation.