Optimizing Query Evaluations Using Reinforcement Learning for Web Search
Open accessThe DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Abstract
In web search, typically a candidate generation step selects a small set of documents---from collections containing as many as billions of web pages---that are subsequently ranked and pruned before being presented to the user. In Bing, the candidate generation involves scanning the index using statically designed match plans that prescribe sequences of different match criteria and stopping conditions. In this work, we pose match planning as a reinforcement learning task and observe up to 20% reduction in index blocks accessed, with small or no degradation in the quality of the candidate sets.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Efficient query evaluation using a two-level retrieval process
Broder, A. Z., Carmel, D., Herscovici, M., Soffer, A., & Zien, J. (2003) · Proceedings of the twelfth international conference on Information and knowledge management · doi:10.1145/956863.956944 · Cited by 407 · cited 1×
“search. Even a 100ms latency has been shown to invoke negative user reactions [2, 18]. A large body of work in information retrieval (IR) has, therefore, focused on efficient query evaluations—e.g., [1, 6, 7]. In the context of machine learning based approaches to retrieval, models have been proposed that incorporate…”
BackgroundEstimating search engine index size variability: a 9-year longitudinal study
van den Bosch, A., Bogers, T., & de Kunder, M. (2016) · Scientometrics · doi:10.1007/s11192-016-1863-z · Cited by 77
Cited by
An Introduction to Neural Information Retrieval
Mitra, B., & Craswell, N. (2018) · Foundations and Trends® in Information Retrieval · doi:10.1561/1500000061 · Cited by 317
Self-citationOverview of the TREC 2020 Deep Learning Track
Craswell, N., Mitra, B., Yilmaz, E., & Campos, D. (2021) · arXiv · doi:10.48550/arxiv.2102.07662 · Cited by 107
Self-citationOverview of the TREC 2019 deep learning track
Craswell, N., Mitra, B., Yilmaz, E., Campos, D., & Voorhees, E. M. (2020) · arXiv · doi:10.48550/arxiv.2003.07820 · Cited by 68
“To analyse if any of the fullrank runs recall more relevant candidates in phase 1 compared to those provided for the reranking subtask, we also report Normalized Cumulative Gain (NCG) [Rosset et al., 2018] at rank 100 and 1000 for the document and passage ranking tasks, respectively.”
MethodSelf-citationTREC Deep Learning Track: Reusable Test Collections in the Large Data Regime
Craswell, N., Mitra, B., Yilmaz, E., Campos, D., Voorhees, E. M., & Soboroff, I. (2021) · Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3404835.3463249 · Cited by 58
“In terms of evaluation metrics, the Deep Learning Track has consistently used NDCG@10 [13] as the primary evaluation metric and NCG [26] (at cutoff 100 and 1000 for the document and passage ranking tasks, respectively) for recall-oriented measure, which we also recommend as the main metrics to the users of the dataset…”
MethodSelf-citationA reinforcement learning framework for relevance feedback
Montazeralghaem, A., Zamani, H., & Allan, J. (2020) · Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3397271.3401099 · Cited by 36
Metadata
- Authors
- Rosset, C., Jose, D., Ghosh, G., Mitra, B., & Tiwary, S.
- Venue
- The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval
- Publisher
- ACM
- Published
- 2018-06-27
- Pages
- 1193-1196
- OpenAlex
- W2797868320
- Cited by
- 31
- References
- 2
- FWCI
- 4.58
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Optimizing Query Evaluations Using Reinforcement Learning for Web Search. https://app.cochat.ai/papers/10.1145/3209978.3210127
- The paper
- Rosset, C., Jose, D., Ghosh, G., Mitra, B., & Tiwary, S. (2018). Optimizing Query Evaluations Using Reinforcement Learning for Web Search. The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. https://doi.org/10.1145/3209978.3210127
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.