Draft:Product quantization
Submission declined on 7 June 2025 by CoconutOctopus (talk).
Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
|  |
 | This is a draft article. It is a work in progress open to editing by anyone. Please ensure core content policies are met before publishing it as a live Wikipedia article. Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL Last edited by CoconutOctopus (talk | contribs) 3 seconds ago. (Update)
Finished drafting? or |
Product quantization (PQ) is a technique for approximate nearest neighbor search.[1] It works by lossily compressing vectors by representing them as a Cartesian product of low-dimensional subspaces and quantizing each subspace independently.[2] Distances can be efficiently computed between product-quantized vectors and an unquantized vector by creating a lookup table, so product quantization can save compute, storage and memory bandwidth.[3]
Product quantization can be used by itself or as a component of more complex ANN search algorithms.[4] The combination of product quantization with inverted files is sometimes known as IVFDAC (inverted file system with asymmetric distance computation): this involves dividing the database into coarse buckets and, for a given query, doing distance computations only with the buckets nearest to that query.[5]
Optimized product quantization (OPQ)
[edit]Optimized product quantization is a widely used enhancement which applies a rotation matrix before quantizing vectors, in order to better take into account the data distribution.[5]
References
[edit]- ^ Jégou, Herve; Douze, Matthijs; Schmid, Cordelia (2010-03-18). "Product Quantization for Nearest Neighbor Search". IEEE Transactions on Pattern Analysis and Machine Intelligence. 33 (1): 117–128. doi:10.1109/TPAMI.2010.57. PMID 21088323. Retrieved 13 April 2025.
- ^ Wu, Ze-bin; Yu, Jun-qing (2019-05-18). "Vector quantization: a review". Frontiers of Information Technology & Electronic Engineering. 20 (4): 507–524. doi:10.1631/FITEE.1700833.
- ^ Guo, Ruiqi; Sun, Philip; Lindgren, Erik; Geng, Quan; Simcha, David; Chern, Felix; Kumar, Sanjiv (2019). "Accelerating Large-Scale Inference with Anisotropic Vector Quantization". arXiv:1908.10396 [cs.LG].
- ^ Douze, Matthijs; Guzhva, Alexandr; Deng, Chengqi; Johnson, Jeff; Szilvasy, Gergely; Mazaré, Pierre-Emmanuel; Lomeli, Maria; Hosseini, Lucas; Jégou, Hervé (2024). "The Faiss library". arXiv:2401.08281 [cs.LG].
- ^ a b Matsui, Yusuke; Uchida, Yusuke; Jégou, Hervé; Satoh, Shin'ichi (2018). "A Survey of Product Quantization". ITE Transactions on Media Technology and Applications. 6 (1): 2–10. doi:10.3169/mta.6.2. Retrieved 13 April 2025.
- in-depth (not just passing mentions about the subject)
- reliable
- secondary
- independent of the subject
Make sure you add references that meet these criteria before resubmitting. Learn about mistakes to avoid when addressing this issue. If no additional references exist, the subject is not suitable for Wikipedia.