Corpus ID: 14219841

Improving Bilayer Product Quantization for Billion-Scale Approximate Nearest Neighbors in High Dimensions

@article{Babenko2014ImprovingBP,
  title={Improving Bilayer Product Quantization for Billion-Scale Approximate Nearest Neighbors in High Dimensions},
  author={Artem Babenko and Victor S. Lempitsky},
  journal={ArXiv},
  year={2014},
  volume={abs/1404.1831}
}
  • Artem Babenko, Victor S. Lempitsky
  • Published in ArXiv 2014
  • Computer Science
  • The top-performing systems for billion-scale high-dimensional approximate nearest neighbor (ANN) search are all based on two-layer architectures that include an indexing structure and a compressed datapoints layer. An indexing structure is crucial as it allows to avoid exhaustive search, while the lossy data compression is needed to fit the dataset into RAM. Several of the most successful systems use product quantization (PQ) for both the indexing and the dataset compression layers. These… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 19 CITATIONS

    Sparse composite quantization

    VIEW 4 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    PQTable: Fast Exact Asymmetric Distance Neighbor Search for Product Quantization Using Hash Tables

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    The Inverted Multi-Index Artem Babenko and

    VIEW 10 EXCERPTS
    CITES BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    The Inverted Multi-Index

    VIEW 10 EXCERPTS
    CITES BACKGROUND & RESULTS

    Composite Quantization

    Mean-Removed Product Quantization for Approximate Nearest Neighbor Search

    RobustiQ: A Robust ANN Search Method for Billion-scale Similarity Search on GPUs

    VIEW 3 EXCERPTS
    CITES BACKGROUND

    Vector and Line Quantization for Billion-scale Similarity Search on GPUs

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    A Biresolution Spectral Framework for Product Quantization

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 14 REFERENCES

    Optimized Product Quantization

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    The inverted multi-index

    VIEW 10 EXCERPTS

    Product Quantization for Nearest Neighbor Search

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Searching in one billion vectors: Re-rank with source coding

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Cartesian K-Means

    VIEW 2 EXCERPTS

    Joint Inverted Indexing

    VIEW 1 EXCERPT

    Optimized Product Quantization for Approximate Nearest Neighbor Search

    VIEW 1 EXCERPT