@inproceedings{10.1145/3512527.3531408,
author = {Amara, Kenza and Douze, Matthijs and Sablayrolles, Alexandre and J\'{e}gou, Herv\'{e}},
title = {Nearest Neighbor Search with Compact Codes: A Decoder Perspective},
year = {2022},
isbn = {9781450392389},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3512527.3531408},
doi = {10.1145/3512527.3531408},
abstract = {Modern approaches for fast retrieval of similar vectors on billion-scaled datasets rely on compressed-domain approaches such as binary sketches or product quantization. These methods minimize a certain loss, typically the Mean Squared Error or other objective functions tailored to the retrieval problem. In this paper, we re-interpret popular methods such as binary hashing or product quantizers as auto-encoders, and point out that they implicitly make suboptimal assumptions on the form of the decoder. We design backward-compatible decoders that improve the reconstruction of the vectors from the same codes, which translates to a better performance in nearest neighbor search. Our method significantly improves over binary hashing methods and product quantization on popular benchmarks.},
booktitle = {Proceedings of the 2022 International Conference on Multimedia Retrieval},
pages = {167–175},
numpages = {9},
keywords = {nearest-neighbors, indexing, quantization},
location = {Newark, NJ, USA},
series = {ICMR '22}
}