Adversarial Autoencoders with Constant-Curvature Latent Manifolds

12/11/2018
by   Daniele Grattarola, et al.
0

Constant-curvature Riemannian manifolds (CCMs) have been shown to be ideal embedding spaces in many application domains, as their non-Euclidean geometry can naturally account for some relevant properties of data, like hierarchy and circularity. In this work, we introduce the CCM adversarial autoencoder (CCM-AAE), a probabilistic generative model trained to represent a data distribution on a CCM. Our method works by matching the aggregated posterior of the CCM-AAE with a probability distribution defined on a CCM, so that the encoder implicitly learns to represent the data on the CCM in order to fool a discriminator network. The geometrical constraint is also explicitly imposed by jointly training the CCM-AAE to maximise the membership degree of the embeddings to the CCM. While several works in recent literature make use of either hyperspherical or hyperbolic manifolds for different learning tasks, ours is the first unified framework to seamlessly deal with CCMs of different curvatures. We show the effectiveness of our model on three different datasets characterised by non-trivial geometry: semi-supervised classification on MNIST, link prediction on two popular citation datasets, and graph-based molecule generation using the QM9 chemical database. Results show that our model compares favourably to other autoencoders based on Euclidean and non-Euclidean geometries on all tasks taken into account.

READ FULL TEXT
research
11/19/2019

Mixed-curvature Variational Autoencoders

It has been shown that using geometric spaces with non-zero curvature in...
research
02/20/2020

Computationally Tractable Riemannian Manifolds for Graph Embeddings

Representing graphs as sets of node embeddings in certain curved Riemann...
research
05/16/2018

Learning Graph Embeddings on Constant-Curvature Manifolds for Change Detection in Graph Streams

The space of graphs is characterized by a non-trivial geometry, which of...
research
11/18/2015

Adversarial Autoencoders

In this paper, we propose the "adversarial autoencoder" (AAE), which is ...
research
10/29/2021

Most probable paths for anisotropic Brownian motions on manifolds

Brownian motion on manifolds with non-trivial diffusion coefficient can ...
research
08/22/2022

Semi-Supervised Manifold Learning with Complexity Decoupled Chart Autoencoders

Autoencoding is a popular method in representation learning. Conventiona...
research
10/09/2020

Conformal retrofitting via Riemannian manifolds: distilling task-specific graphs into pretrained embeddings

Pretrained (language) embeddings are versatile, task-agnostic feature re...

Please sign up or login with your details

Forgot password? Click here to reset