English

Learning List-Level Domain-Invariant Representations for Ranking

Information Retrieval 2023-11-01 v3 Artificial Intelligence Computation and Language Machine Learning

Abstract

Domain adaptation aims to transfer the knowledge learned on (data-rich) source domains to (low-resource) target domains, and a popular method is invariant representation learning, which matches and aligns the data distributions on the feature space. Although this method is studied extensively and applied on classification and regression problems, its adoption on ranking problems is sporadic, and the few existing implementations lack theoretical justifications. This paper revisits invariant representation learning for ranking. Upon reviewing prior work, we found that they implement what we call item-level alignment, which aligns the distributions of the items being ranked from all lists in aggregate but ignores their list structure. However, the list structure should be leveraged, because it is intrinsic to ranking problems where the data and the metrics are defined and computed on lists, not the items by themselves. To close this discrepancy, we propose list-level alignment -- learning domain-invariant representations at the higher level of lists. The benefits are twofold: it leads to the first domain adaptation generalization bound for ranking, in turn providing theoretical support for the proposed method, and it achieves better empirical transfer performance for unsupervised domain adaptation on ranking tasks, including passage reranking.

Keywords

Cite

@article{arxiv.2212.10764,
  title  = {Learning List-Level Domain-Invariant Representations for Ranking},
  author = {Ruicheng Xian and Honglei Zhuang and Zhen Qin and Hamed Zamani and Jing Lu and Ji Ma and Kai Hui and Han Zhao and Xuanhui Wang and Michael Bendersky},
  journal= {arXiv preprint arXiv:2212.10764},
  year   = {2023}
}

Comments

NeurIPS 2023. Comparison to v1: revised presentation and proof of Corollary 4.9

R2 v1 2026-06-28T07:46:05.555Z