Geometrically Enriched Latent Spaces

08/02/2020
by   Georgios Arvanitidis, et al.
20

A common assumption in generative models is that the generator immerses the latent space into a Euclidean ambient space. Instead, we consider the ambient space to be a Riemannian manifold, which allows for encoding domain knowledge through the associated Riemannian metric. Shortest paths can then be defined accordingly in the latent space to both follow the learned manifold and respect the ambient geometry. Through careful design of the ambient metric we can ensure that shortest paths are well-behaved even for deterministic generators that otherwise would exhibit a misleading bias. Experimentally we show that our approach improves interpretability of learned representations both using stochastic and deterministic generators.

READ FULL TEXT

page 7

page 8

page 16

page 19

page 20

page 21

page 23

research
12/20/2022

Identifying latent distances with Finslerian geometry

Riemannian geometry provides powerful tools to explore the latent space ...
research
04/26/2022

Double Diffusion Maps and their Latent Harmonics for Scientific Computations in Latent Space

We introduce a data-driven approach to building reduced dynamical models...
research
02/06/2022

Estimating the Euclidean Quantum Propagator with Deep Generative Modelling of Feynman paths

Feynman path integrals provide an elegant, classically-inspired represen...
research
03/09/2021

A prior-based approximate latent Riemannian metric

Stochastic generative models enable us to capture the geometric structur...
research
05/04/2021

Riemannian Geometry with differentiable ambient space and metric operator

We show Riemannian geometry could be studied by identifying the tangent ...
research
09/26/2019

A theory of condition for unconstrained perturbations

Traditionally, the theory of condition numbers assumes errors in the dat...
research
08/20/2019

Expected path length on random manifolds

Manifold learning seeks a low dimensional representation that faithfully...

Please sign up or login with your details

Forgot password? Click here to reset