Matches in SemOpenAlex for { <https://semopenalex.org/work/W4377157057> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4377157057 abstract "The goal of unbiased learning to rank (ULTR) is to leverage implicit user feedback for optimizing learning-to-rank systems. Among existing solutions, automatic ULTR algorithms that jointly learn user bias models (i.e., propensity models) with unbiased rankers have received a lot of attention due to their superior performance and low deployment cost in practice. Despite their theoretical soundness, the effectiveness is usually justified under a weak logging policy, where the ranking model can barely rank documents according to their relevance to the query. However, when the logging policy is strong, e.g., an industry-deployed ranking policy, the reported effectiveness cannot be reproduced. In this paper, we first investigate ULTR from a causal perspective and uncover a negative result: existing ULTR algorithms fail to address the issue of propensity overestimation caused by the query-document relevance confounder. Then, we propose a new learning objective based on backdoor adjustment and highlight its differences from conventional propensity models, which reveal the prevalence of propensity overestimation. On top of that, we introduce a novel propensity model called Logging-Policy-aware Propensity (LPP) model and its distinctive two-step optimization strategy, which allows for the joint learning of LPP and ranking models within the automatic ULTR framework, and actualize the unconfounded propensity estimation for ULTR. Extensive experiments on two benchmarks demonstrate the effectiveness and generalizability of the proposed method." @default.
- W4377157057 created "2023-05-21" @default.
- W4377157057 creator A5006671485 @default.
- W4377157057 creator A5013312078 @default.
- W4377157057 creator A5042328810 @default.
- W4377157057 creator A5054482111 @default.
- W4377157057 creator A5065789784 @default.
- W4377157057 creator A5089307887 @default.
- W4377157057 creator A5089655391 @default.
- W4377157057 date "2023-05-16" @default.
- W4377157057 modified "2023-09-30" @default.
- W4377157057 title "Unconfounded Propensity Estimation for Unbiased Ranking" @default.
- W4377157057 doi "https://doi.org/10.48550/arxiv.2305.09918" @default.
- W4377157057 hasPublicationYear "2023" @default.
- W4377157057 type Work @default.
- W4377157057 citedByCount "0" @default.
- W4377157057 crossrefType "posted-content" @default.
- W4377157057 hasAuthorship W4377157057A5006671485 @default.
- W4377157057 hasAuthorship W4377157057A5013312078 @default.
- W4377157057 hasAuthorship W4377157057A5042328810 @default.
- W4377157057 hasAuthorship W4377157057A5054482111 @default.
- W4377157057 hasAuthorship W4377157057A5065789784 @default.
- W4377157057 hasAuthorship W4377157057A5089307887 @default.
- W4377157057 hasAuthorship W4377157057A5089655391 @default.
- W4377157057 hasBestOaLocation W43771570571 @default.
- W4377157057 hasConcept C105795698 @default.
- W4377157057 hasConcept C114614502 @default.
- W4377157057 hasConcept C119857082 @default.
- W4377157057 hasConcept C124101348 @default.
- W4377157057 hasConcept C149782125 @default.
- W4377157057 hasConcept C153083717 @default.
- W4377157057 hasConcept C154945302 @default.
- W4377157057 hasConcept C158154518 @default.
- W4377157057 hasConcept C164226766 @default.
- W4377157057 hasConcept C17744445 @default.
- W4377157057 hasConcept C17923572 @default.
- W4377157057 hasConcept C189430467 @default.
- W4377157057 hasConcept C199539241 @default.
- W4377157057 hasConcept C33923547 @default.
- W4377157057 hasConcept C41008148 @default.
- W4377157057 hasConcept C86037889 @default.
- W4377157057 hasConceptScore W4377157057C105795698 @default.
- W4377157057 hasConceptScore W4377157057C114614502 @default.
- W4377157057 hasConceptScore W4377157057C119857082 @default.
- W4377157057 hasConceptScore W4377157057C124101348 @default.
- W4377157057 hasConceptScore W4377157057C149782125 @default.
- W4377157057 hasConceptScore W4377157057C153083717 @default.
- W4377157057 hasConceptScore W4377157057C154945302 @default.
- W4377157057 hasConceptScore W4377157057C158154518 @default.
- W4377157057 hasConceptScore W4377157057C164226766 @default.
- W4377157057 hasConceptScore W4377157057C17744445 @default.
- W4377157057 hasConceptScore W4377157057C17923572 @default.
- W4377157057 hasConceptScore W4377157057C189430467 @default.
- W4377157057 hasConceptScore W4377157057C199539241 @default.
- W4377157057 hasConceptScore W4377157057C33923547 @default.
- W4377157057 hasConceptScore W4377157057C41008148 @default.
- W4377157057 hasConceptScore W4377157057C86037889 @default.
- W4377157057 hasLocation W43771570571 @default.
- W4377157057 hasLocation W43771570572 @default.
- W4377157057 hasOpenAccess W4377157057 @default.
- W4377157057 hasPrimaryLocation W43771570571 @default.
- W4377157057 hasRelatedWork W104148947 @default.
- W4377157057 hasRelatedWork W1786507113 @default.
- W4377157057 hasRelatedWork W2104465941 @default.
- W4377157057 hasRelatedWork W2293317945 @default.
- W4377157057 hasRelatedWork W2311112415 @default.
- W4377157057 hasRelatedWork W2798835721 @default.
- W4377157057 hasRelatedWork W4318960487 @default.
- W4377157057 hasRelatedWork W4323349240 @default.
- W4377157057 hasRelatedWork W4378713513 @default.
- W4377157057 hasRelatedWork W614547436 @default.
- W4377157057 isParatext "false" @default.
- W4377157057 isRetracted "false" @default.
- W4377157057 workType "article" @default.