Pairwise Quantization

06/05/2016
by   Artem Babenko, et al.
0

We consider the task of lossy compression of high-dimensional vectors through quantization. We propose the approach that learns quantization parameters by minimizing the distortion of scalar products and squared distances between pairs of points. This is in contrast to previous works that obtain these parameters through the minimization of the reconstruction error of individual points. The proposed approach proceeds by finding a linear transformation of the data that effectively reduces the minimization of the pairwise distortions to the minimization of individual reconstruction errors. After such transformation, any of the previously-proposed quantization approaches can be used. Despite the simplicity of this transformation, the experiments demonstrate that it achieves considerable reduction of the pairwise distortions compared to applying quantization directly to the untransformed data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/05/2016

Towards the Limit of Network Quantization

Network quantization is one of network compression techniques to reduce ...
research
12/11/2021

Piercing Pairwise Intersecting Geodesic Disks by Five Points

Given a simple polygon P on n vertices, and a set D of m pairwise inters...
research
02/10/2022

Quantization in Layer's Input is Matter

In this paper, we will show that the quantization in layer's input is mo...
research
07/31/2022

Symmetry Regularization and Saturating Nonlinearity for Robust Quantization

Robust quantization improves the tolerance of networks for various imple...
research
12/18/2019

Interleaved Composite Quantization for High-Dimensional Similarity Search

Similarity search retrieves the nearest neighbors of a query vector from...
research
07/08/2019

Multi-Scale Vector Quantization with Reconstruction Trees

We propose and study a multi-scale approach to vector quantization. We d...
research
07/07/2017

A multi-layer image representation using Regularized Residual Quantization: application to compression and denoising

A learning-based framework for representation of domain-specific images ...

Please sign up or login with your details

Forgot password? Click here to reset