code for "Continuous Hierarchical Representations with Poincaré Variational Auto-Encoders".
-
Updated
Oct 3, 2023 - Python
code for "Continuous Hierarchical Representations with Poincaré Variational Auto-Encoders".
Taxonomy refinement method to improve domain-specific taxonomy systems.
Hyperbolic Busemann Learning with Ideal Prototypes, NeurIPS2021
Train poincare embedding using gensim
K-Means algorithm in the Poincare Disk Model
CS-663 Course Project
A "navigator" for exploring 2-dimensional hyperbolic space using the Poincaré disc model. The entire graph or point configuration can be translated and rotated by dragging the ambient, and individual points can be moved, too.
Hierarchical Book Recommendations
Hyperbolic embedding using AudioSet ontology.
Multi-Relational Hyperbolic Word Embeddings from Natural Language Definitions
Moteur de compression de graphes par plongement hyperbolique. Transforme le stockage quadratique O(N²) en linéaire O(N) et accélère la recherche de 100M× (O(log N)). Implémentation Pure Python
Research-backed MCP system using Arabic morphology, transformer embeddings, and Poincaré geometry to ground semantic relationships in classical Arabic.
Implementation of Poincaré embeddings in PyTorch for learning hierarchical representations in hyperbolic space.
A hyperbolic context manifold, built and tested: Poincare disks vs flat space for hierarchy, over-smoothing and link prediction. Code for the essay Escaping Flatland.
To associate your repository with the poincare-embeddings topic, visit your repo's landing page and select "manage topics."