An Alternative Cross Entropy Loss For Learning-to-rank

Abstract

Listwise learning-to-rank methods form a powerful class of ranking algorithms that are widely adopted in applications such as information retrieval. These algorithms learn to rank a set of items by optimizing a loss that is a function of the entire set -- as a surrogate to a typically non-differentiable ranking metric. Despite their empirical success, existing listwise methods are based on heuristics and remain theoretically ill-understood. In particular, none of the empirically successful loss functions are related to ranking metrics. In this work, we propose a cross entropy-based learning-to-rank loss function that is theoretically sound, is a convex bound on NDCG -- a popular ranking metric -- and is consistent with NDCG under learning scenarios common in information retrieval. Furthermore, empirical evaluation of an implementation of the proposed method with gradient boosting machines on benchmark learning-to-rank datasets demonstrates the superiority of our proposed formulation ov

Stats

• citations34
• S2 citations—
• github stars0
• HF likes0
• heat score11.58
• arxiv keybruch2019an

Related papers

• Optimization Of Rank Losses For Image Retrieval (2023)9.43
• Croloss: Towards A Customizable Loss For Retrieval Models In Recommender Systems (2022)5.84
• Expertrank: A Multi-level Coarse-grained Expert-based Listwise Ranking Loss (2021)0.00
• Towards Non-parametric Learning To Rank (2018)0.00
• Ranked List Loss For Deep Metric Learning (2019)7.50
• Instance Cross Entropy For Deep Metric Learning (2019)0.00
• Setrank: Learning A Permutation-invariant Ranking Model For Information Retrieval (2019)0.00
• Hashing As Tie-aware Learning To Rank (2017)13.11