A neural network catalyzer for multi-dimensional similarity search

06/08/2018
by   Alexandre Sablayrolles, et al.
0

This paper aims at learning a function mapping input vectors to an output space in a way that improves high-dimensional similarity search. As a proxy objective, we design and train a neural network that favors uniformity in the spherical output space, while preserving the neighborhood structure after the mapping. For this purpose, we propose a new regularizer derived from the Kozachenko-Leonenko differential entropy estimator and combine it with a locality-aware triplet loss. Our method operates as a catalyzer for traditional indexing methods such as locality sensitive hashing or iterative quantization, boosting the overall recall. Additionally, the network output distribution makes it possible to leverage structured quantizers with efficient algebraic encoding, in particular spherical lattice quantizers such as the Gosset lattice E8. Our experiments show that this approach is competitive with state-of-the-art methods such as optimized product quantization.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/05/2018

Improving Similarity Search with High-dimensional Locality-sensitive Hashing

We propose a new class of data-independent locality-sensitive hashing (L...
research
10/05/2022

Active Image Indexing

Image copy detection and retrieval from large databases leverage two com...
research
06/07/2019

Deep Spherical Quantization for Image Search

Hashing methods, which encode high-dimensional images with compact discr...
research
10/06/2011

Bayesian Locality Sensitive Hashing for Fast Similarity Search

Given a collection of objects and an associated similarity measure, the ...
research
04/08/2019

Subsets and Supermajorities: Unifying Hashing-based Set Similarity Search

We consider the problem of designing Locality Sensitive Filters (LSF) fo...
research
12/26/2012

Hyperplane Arrangements and Locality-Sensitive Hashing with Lift

Locality-sensitive hashing converts high-dimensional feature vectors, su...
research
06/15/2020

Iterative Nadaraya-Watson Distribution Transfer for Colour Grading

We propose a new method with Nadaraya-Watson that maps one N-dimensional...

Please sign up or login with your details

Forgot password? Click here to reset