← Authors

Gary Becigneul

9 papers · 2018–2021 · 5 conferences · across top CS/AI conferences

Achievements

πŸ—ΊοΈ Taxonomy Completionist (12) 🧭 Keyword Pioneer πŸŒ‰ Interdisciplinary Bridge 🌍 Conference Polyglot (5) 🐣 Hot Topic Early Bird 🐝 Cross-Pollinator (11) 🌈 Renaissance Researcher (5) πŸ† Grand Slam

Conferences

ICML (3) AISTATS (2) ICLR (2) AAAI (1) NIPS (1)

Top co-authors

Octavian Ganea (4) Octavian-Eugen Ganea (3) Aurelien Lucchi (2) Thomas Hofmann (2) Foivos Alimisis (2) Antonio Orvieto (2) Ondrej Skopek (1) Gregor Bachmann (1) Calin Cruceru (1) Sylvain Gelly (1)

Keywords

riemannian manifold (2) matrix factorization (2) hyperbolic geometry (2) convex optimization (1) graph representation (1) graph embedding (1) accelerated gradient (1) recurrent neural network (1) language model (1) node classification (1) hierarchical embedding (1) first-order method (1) directed acyclic graph (1) nesterov acceleration (1) hyperbolic embedding (1) cross entropy (1) hyperbolic space (1) sentence embedding (1) geodesic distance (1) monotonic function (1)

Papers

Computationally Tractable Riemannian Manifolds for Graph Embeddings AAAI 2021
Momentum Improves Optimization on Riemannian Manifolds AISTATS 2021
Mixed-curvature Variational Autoencoders ICLR 2020
A Continuous-time Perspective for Modeling Acceleration in Riemannian Optimization AISTATS 2020
Constant Curvature Graph Convolutional Networks ICML 2020
Riemannian Adaptive Optimization Methods ICLR 2019
Breaking the Softmax Bottleneck via Learnable Monotonic Pointwise Non-linearities ICML 2019
Hyperbolic Neural Networks NIPS 2018
Hyperbolic Entailment Cones for Learning Hierarchical Embeddings ICML 2018