Parallel Total Variation Distance Estimation with Neural Networks for Merging Over-Clusterings

12/09/2019
by   Christian Reiser, et al.
0

We consider the initial situation where a dataset has been over-partitioned into k clusters and seek a domain independent way to merge those initial clusters. We identify the total variation distance (TVD) as suitable for this goal. By exploiting the relation of the TVD to the Bayes accuracy we show how neural networks can be used to estimate TVDs between all pairs of clusters in parallel. Crucially, the needed memory space is decreased by reducing the required number of output neurons from k^2 to k. On realistically obtained over-clusterings of ImageNet subsets it is demonstrated that our TVD estimates lead to better merge decisions than those obtained by relying on state-of-the-art unsupervised representations. Further the generality of the approach is verified by evaluating it on a a point cloud dataset.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/25/2014

Continuum limit of total variation on point clouds

We consider point clouds obtained as random samples of a measure on a Eu...
research
07/03/2016

Variational limits of k-NN graph based functionals on data clouds

We consider i.i.d. samples x_1, ..., x_n from a measure ν with density s...
research
09/16/2023

Quantum Pseudorandom Scramblers

Quantum pseudorandom state generators (PRSGs) have stimulated exciting d...
research
12/14/2022

Total variation distance between a jump-equation and its Gaussian approximation

We deal with stochastic differential equations with jumps. In order to o...
research
09/25/2009

Discrete MDL Predicts in Total Variation

The Minimum Description Length (MDL) principle selects the model that ha...
research
11/16/2015

Probabilistic Segmentation via Total Variation Regularization

We present a convex approach to probabilistic segmentation and modeling ...
research
07/03/2023

Interpolation of Point Distributions for Digital Stippling

We present a new way to merge any two point distribution approaches usin...

Please sign up or login with your details

Forgot password? Click here to reset