Individualized non-uniform quantization for vector search

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Tepper, Mariano, Willke, Ted
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909801934487552
author Tepper, Mariano
Willke, Ted
author_facet Tepper, Mariano
Willke, Ted
contents Embedding vectors are widely used for representing unstructured data and searching through it for semantically similar items. However, the large size of these vectors, due to their high-dimensionality, creates problems for modern vector search techniques: retrieving large vectors from memory/storage is expensive and their footprint is costly. In this work, we present NVQ (non-uniform vector quantization), a new vector compression technique that is computationally and spatially efficient in the high-fidelity regime. The core in NVQ is to use novel parsimonious and computationally efficient nonlinearities for building non-uniform vector quantizers. Critically, these quantizers are \emph{individually} learned for each indexed vector. Our experimental results show that NVQ exhibits improved accuracy compared to the state of the art with a minimal computational cost.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18471
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Individualized non-uniform quantization for vector search
Tepper, Mariano
Willke, Ted
Machine Learning
Information Retrieval
Embedding vectors are widely used for representing unstructured data and searching through it for semantically similar items. However, the large size of these vectors, due to their high-dimensionality, creates problems for modern vector search techniques: retrieving large vectors from memory/storage is expensive and their footprint is costly. In this work, we present NVQ (non-uniform vector quantization), a new vector compression technique that is computationally and spatially efficient in the high-fidelity regime. The core in NVQ is to use novel parsimonious and computationally efficient nonlinearities for building non-uniform vector quantizers. Critically, these quantizers are \emph{individually} learned for each indexed vector. Our experimental results show that NVQ exhibits improved accuracy compared to the state of the art with a minimal computational cost.
title Individualized non-uniform quantization for vector search
topic Machine Learning
Information Retrieval
url https://arxiv.org/abs/2509.18471