TMI: Thermodynamic inference of data manifolds

11/21/2019
by   Purushottam D. Dixit, et al.
0

The Gibbs-Boltzmann distribution offers a physically interpretable way to massively reduce the dimensionality of high dimensional probability distributions where the extensive variables are `features' and the intensive variables are `descriptors'. However, not all probability distributions can be modeled using the Gibbs-Boltzmann form. Here, we present TMI: TMI, Thermodynamic Manifold Inference; a thermodynamic approach to approximate a collection of arbitrary distributions. TMI simultaneously learns from data intensive and extensive variables and achieves dimensionality reduction through a multiplicative, positive valued, and interpretable decomposition of the data. Importantly, the reduced dimensional space of intensive parameters is not homogeneous. The Gibbs-Boltzmann distribution defines an analytically tractable Riemannian metric on the space of intensive variables allowing us to calculate geodesics and volume elements. We discuss the applications of TMI with multiple real and artificial data sets. Possible extensions are discussed as well.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/20/2017

Boltzmann machines and energy-based models

We review Boltzmann machines and energy-based models. A Boltzmann machin...
research
05/18/2021

Parametrization invariant interpretation of priors and posteriors

In this paper we leverage on probability over Riemannian manifolds to re...
research
11/13/2018

Sampling from manifold-restricted distributions using tangent bundle projections

A common problem in Bayesian inference is the sampling of target probabi...
research
07/17/2020

Parameter estimation for Gibbs distributions

We consider Gibbs distributions, which are families of probability distr...
research
06/02/2021

Rectangular Flows for Manifold Learning

Normalizing flows are invertible neural networks with tractable change-o...
research
02/08/2019

Can Genetic Programming Do Manifold Learning Too?

Exploratory data analysis is a fundamental aspect of knowledge discovery...
research
02/05/2015

A Confident Information First Principle for Parametric Reduction and Model Selection of Boltzmann Machines

Typical dimensionality reduction (DR) methods are often data-oriented, f...

Please sign up or login with your details

Forgot password? Click here to reset