Nowadays, state-of-the-art learning-to-rank methods are based on gradient-boosted decision trees (GBDT). The most well-known algorithm is LambdaMART which was proposed more than a decade ago. Recently, several other GBDT-based ranking algorithms were proposed. In this paper, we thoroughly analyze these methods in a unified setup. In particular, we address the following questions. Is direct optimization of a smoothed ranking loss preferable over optimizing a convex surrogate? How to properly construct and smooth surrogate ranking losses? To address these questions, we compare LambdaMART with YetiRank and StochasticRank methods and their modifications. We also propose a simple improvement of the YetiRank approach that allows for optimizing specific ranking loss functions. As a result, we gain insights into learning-to-rank techniques and obtain a new state-of-the-art algorithm.

Ivan Lyzhin

Aleksei Ustimenko

Andrey Gulin

Liudmila Prokhorenkova

Learning to rank is a central problem in information retrieval. The objective is to
rank a given set of items to optimize the overall utility of the list.

Ranking

Gradient boosting iteratively combines weak learners (usually decision trees) to create a stronger model. It achieves state-of-the-art results on tabular data with heterogeneous features.

Which Tricks Are Important for Learning to Rank?