Connectivity-Optimized Representation Learning via Persistent Homology

Abstract

We study the problem of learning representations with controllable connectivity properties. This is beneficial in situations when the imposed structure can be leveraged upstream. In particular, we control the connectivity of an autoencoder's latent space via a novel type of loss, operating on information from persistent homology. Under mild conditions, this loss is differentiable and we present a theoretical analysis of the properties induced by the loss. We choose one-class learning as our upstream task and demonstrate that the imposed structure enables informed parameter selection for modeling the in-class distribution via kernel density estimators. Evaluated on computer vision data, these one-class models exhibit competitive performance and, in a low sample size regime, outperform other methods by a large margin. Notably, our results indicate that a single autoencoder, trained on auxiliary (unlabeled) data, yields a mapping into latent space that can be reused across datasets for one-class learning.

Cited by in corpus (17)

• A Topological Loss Function for Deep-Learning based Image Segmentation using Persistent Homology
• A persistent homology-based topological loss for CNN-based multi-class segmentation of CMR
• Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation
• Uncovering the Topology of Time-Varying fMRI Data using Cubical Persistence
• Optimizing persistent homology based functions
• Path homologies of deep feedforward networks
• Topology-Aware Segmentation Using Discrete Morse Theory
• Activation Landscapes as a Topological Summary of Neural Network Performance
• Localization in the Crowd with Topological Constraints
• Trigger Hunting with a Topological Prior for Trojan Detection
• PLLay: Efficient Topological Layer based on Persistence Landscapes
• Topological Information Retrieval with Dilation-Invariant Bottleneck Comparative Measures
• Intrinsic Dimension, Persistent Homology and Generalization in Neural Networks
• Topological Detection of Trojaned Neural Networks
• Some Geometrical and Topological Properties of DNNs' Decision Boundaries
• Graph Filtration Learning
• Topologically Densified Distributions