Semi-supervised Ranking Pursuit

TL;DR

This paper introduces a semi-supervised ranking algorithm that efficiently learns sparse utility functions, outperforming existing methods by leveraging unlabeled data and providing sparser solutions.

Contribution

It presents a novel semi-supervised ranking algorithm based on a generalized kernel matching pursuit, capable of handling ranking and regression tasks with efficient search for multiple solutions.

Findings

Outperforms state-of-the-art methods with unlabeled data

Provides sparser solutions in supervised scenarios

Demonstrates effectiveness in ranking and regression tasks

Abstract

We propose a novel sparse preference learning/ranking algorithm. Our algorithm approximates the true utility function by a weighted sum of basis functions using the squared loss on pairs of data points, and is a generalization of the kernel matching pursuit method. It can operate both in a supervised and a semi-supervised setting and allows efficient search for multiple, near-optimal solutions. Furthermore, we describe the extension of the algorithm suitable for combined ranking and regression tasks. In our experiments we demonstrate that the proposed algorithm outperforms several state-of-the-art learning methods when taking into account unlabeled data and performs comparably in a supervised learning scenario, while providing sparser solutions.