Accelerating Large-Scale Inference with Anisotropic Vector Quantization

Code

• github.com/google-research/google-research/tree/master/scann
  Officialjax★ 0
• github.com/jbellis/jvector
  none★ 1,695
• github.com/datastax/jvector
  none★ 0
• github.com/AxelvL/AHPQ.jl
  none★ 0
• github.com/loveheaven/scann
  tf★ 0

Abstract

Quantization based techniques are the current state-of-the-art for scaling maximum inner product search to massive databases. Traditional approaches to quantization aim to minimize the reconstruction error of the database points. Based on the observation that for a given query, the database points that have the largest inner products are more relevant, we develop a family of anisotropic quantization loss functions. Under natural statistical assumptions, we show that quantization with these loss functions leads to a new variant of vector quantization that more greatly penalizes the parallel component of a datapoint's residual relative to its orthogonal component. The proposed approach achieves state-of-the-art results on the public benchmarks available at ann-benchmarks.com.

Tasks

Reproductions