English

Spreading vectors for similarity search

Machine Learning 2019-09-02 v3 Machine Learning

Abstract

Discretizing multi-dimensional data distributions is a fundamental step of modern indexing methods. State-of-the-art techniques learn parameters of quantizers on training data for optimal performance, thus adapting quantizers to the data. In this work, we propose to reverse this paradigm and adapt the data to the quantizer: we train a neural net which last layer forms a fixed parameter-free quantizer, such as pre-defined points of a hyper-sphere. As a proxy objective, we design and train a neural network that favors uniformity in the spherical latent space, while preserving the neighborhood structure after the mapping. We propose a new regularizer derived from the Kozachenko--Leonenko differential entropy estimator to enforce uniformity and combine it with a locality-aware triplet loss. Experiments show that our end-to-end approach outperforms most learned quantization methods, and is competitive with the state of the art on widely adopted benchmarks. Furthermore, we show that training without the quantization step results in almost no difference in accuracy, but yields a generic catalyzer that can be applied with any subsequent quantizer.

Keywords

Cite

@article{arxiv.1806.03198,
  title  = {Spreading vectors for similarity search},
  author = {Alexandre Sablayrolles and Matthijs Douze and Cordelia Schmid and Hervé Jégou},
  journal= {arXiv preprint arXiv:1806.03198},
  year   = {2019}
}

Comments

Published at ICLR 2019

R2 v1 2026-06-23T02:23:45.841Z