Conformal Disentanglement: A Neural Framework for Perspective Synthesis and Differentiation

TL;DR

This paper introduces a neural autoencoder framework that disentangles common and sensor-specific information in heterogeneous observations, enabling better synthesis and understanding of multi-perspective data.

Contribution

It proposes a novel neural framework with orthogonality constraints to identify and separate common and uncommon information across diverse sensors.

Findings

Successfully disentangles shared and unique information in multi-sensor data

Improves perspective synthesis by isolating relevant features

Demonstrates effectiveness on multiple computational examples

Abstract

For multiple scientific endeavors it is common to measure a phenomenon of interest in more than one ways. We make observations of objects from several different perspectives in space, at different points in time; we may also measure different properties of a mixture using different types of instruments. After collecting this heterogeneous information, it is necessary to be able to synthesize a complete picture of what is `common' across its sources: the subject we ultimately want to study. However, isolated (`clean') observations of a system are not always possible: observations often contain information about other systems in its environment, or about the measuring instruments themselves. In that sense, each observation may contain information that `does not matter' to the original object of study; this `uncommon' information between sensors observing the same object may still be important, and decoupling it from the main signal(s) useful. We introduce a neural network autoencoder framework capable of both tasks: it is structured to identify `common' variables, and, making use of orthogonality constraints to define geometric independence, to also identify disentangled `uncommon' information originating from the heterogeneous sensors. We demonstrate applications in several computational examples.