Topological pooling on graphs
Graph neural networks (GNNs) have demonstrated a significant success in various graph
learning tasks, from graph classification to anomaly detection. There recently has emerged a …
learning tasks, from graph classification to anomaly detection. There recently has emerged a …
Autoencoders for a manifold learning problem with a jacobian rank constraint
We formulate the manifold learning problem as the problem of finding an operator that maps
any point to a close neighbor that lies on a “hidden” k-dimensional manifold. We call this …
any point to a close neighbor that lies on a “hidden” k-dimensional manifold. We call this …
Geomca: Geometric evaluation of data representations
Evaluating the quality of learned representations without relying on a downstream task
remains one of the challenges in representation learning. In this work, we present Geometric …
remains one of the challenges in representation learning. In this work, we present Geometric …
InvMap and Witness Simplicial Variational Auto-Encoders
Variational auto-encoders (VAEs) are deep generative models used for unsupervised
learning, however their standard version is not topology-aware in practice since the data …
learning, however their standard version is not topology-aware in practice since the data …
When Witnesses Defend: A Witness Graph Topological Layer for Adversarial Graph Learning
Capitalizing on the intuitive premise that shape characteristics are more robust to
perturbations, we bridge adversarial graph learning with the emerging tools from …
perturbations, we bridge adversarial graph learning with the emerging tools from …
Local distance preserving auto-encoders using continuous k-nearest neighbours graphs
N Chen, P van der Smagt, B Cseke - arXiv preprint arXiv:2206.05909, 2022 - arxiv.org
Auto-encoder models that preserve similarities in the data are a popular tool in
representation learning. In this paper we introduce several auto-encoder models that …
representation learning. In this paper we introduce several auto-encoder models that …
Towards Topology-Aware Variational Auto-Encoders: From InvMap-VAE to Witness Simplicial VAE
AA Medbouhi - 2022 - diva-portal.org
Abstract Variational Auto-Encoders (VAEs) are one of the most famous deep generative
models. After showing that standard VAEs may not preserve the topology, that is the shape …
models. After showing that standard VAEs may not preserve the topology, that is the shape …
[PDF][PDF] Local distance preserving auto-encoders using Continuous k-Nearest Neighbours graphs
NCP van der Smagt, B Cseke - arXiv preprint arXiv:2206.05909, 2022 - researchgate.net
Auto-encoder models that preserve similarities in the data are a popular tool in
representation learning. In this paper we introduce several auto-encoder models that …
representation learning. In this paper we introduce several auto-encoder models that …