Octavian-Eugen Ganea

* Submitted to International Conference on Machine Learning (ICML) 2020

Access Paper or Ask Questions

Mixed-curvature Variational Autoencoders

Nov 19, 2019
Ondrej Skopek, Octavian-Eugen Ganea, Gary Bécigneul

Access Paper or Ask Questions

Constant Curvature Graph Convolutional Networks

Nov 12, 2019
Gregor Bachmann, Gary Bécigneul, Octavian-Eugen Ganea

Access Paper or Ask Questions

Noise Contrastive Variational Autoencoders

Jul 31, 2019
Octavian-Eugen Ganea, Yashas Annadani, Gary Bécigneul

* There is a mistake common to all the main proofs. In summary, what we find are saddle points or global maxima of the respective loss functions and not the global minima. We apologize for this

Access Paper or Ask Questions

Breaking the Softmax Bottleneck via Learnable Monotonic Pointwise Non-linearities

Feb 21, 2019
Octavian-Eugen Ganea, Sylvain Gelly, Gary Bécigneul, Aliaksei Severyn

Access Paper or Ask Questions

Poincaré GloVe: Hyperbolic Word Embeddings

Oct 15, 2018
Alexandru Tifrea, Gary Bécigneul, Octavian-Eugen Ganea

Access Paper or Ask Questions

Riemannian Adaptive Optimization Methods

Oct 01, 2018
Gary Bécigneul, Octavian-Eugen Ganea

Access Paper or Ask Questions

Learning and Evaluating Sparse Interpretable Sentence Embeddings

Sep 25, 2018
Valentin Trifonov, Octavian-Eugen Ganea, Anna Potapenko, Thomas Hofmann

* Will be presented at the workshop "Analyzing and interpreting neural networks for NLP", collocated with the EMNLP 2018 conference in Brussels

Access Paper or Ask Questions