Mihaela Rosca

Publication Activity (10 Years)

Years Active: 2016-2024
Publications (10 Years): 27

Top Topics

Gradient Estimation

Reinforcement Learning

Top Venues

Publications

Benoit Dherin, Mihaela Rosca
Corridor Geometry in Gradient-Based Optimization. CoRR (2024)
Piotr Mirowski, David Warde-Farley, Mihaela Rosca, Matthew Koichi Grimes, Yana Hasson, Hyunjik Kim, Mélanie Rey, Simon Osindero, Suman V. Ravuri, Shakir Mohamed
Neural Compression of Atmospheric States. CoRR (2024)
Mihaela Rosca, Marc Peter Deisenroth
Implicit regularisation in stochastic gradient descent: from single-objective to two-player games. CoRR (2023)
Mihaela Rosca, Yan Wu, Chongli Qin, Benoit Dherin
On a continuous time model of gradient descent dynamics and instability in deep learning. CoRR (2023)
Mihaela Rosca, Yan Wu, Chongli Qin, Benoit Dherin
On a continuous time model of gradient descent dynamics and instability in deep learning. Trans. Mach. Learn. Res. 2023 (2023)
Rares Iordan, Marc Peter Deisenroth, Mihaela Rosca
Investigating the Edge of Stability Phenomenon in Reinforcement Learning. CoRR (2023)
Benoit Dherin, Michael Munn, Mihaela Rosca, David G. T. Barrett
Why neural networks find simple solutions: the many regularizers of geometric complexity. CoRR (2022)
Benoit Dherin, Michael Munn, Mihaela Rosca, David Barrett
Why neural networks find simple solutions: The many regularizers of geometric complexity. NeurIPS (2022)
Florin Gogianu, Tudor Berariu, Mihaela Rosca, Claudia Clopath, Lucian Busoniu, Razvan Pascanu
Spectral Normalisation for Deep Reinforcement Learning: an Optimisation Perspective. CoRR (2021)
Mihaela Rosca, Yan Wu, Benoit Dherin, David Barrett
Discretization Drift in Two-Player Games. ICML (2021)
Mihaela Rosca, Yan Wu, Benoit Dherin, David G. T. Barrett
Discretization Drift in Two-Player Games. CoRR (2021)
Florin Gogianu, Tudor Berariu, Mihaela Rosca, Claudia Clopath, Lucian Busoniu, Razvan Pascanu
Spectral Normalisation for Deep Reinforcement Learning: An Optimisation Perspective. ICML (2021)
Shakir Mohamed, Mihaela Rosca, Michael Figurnov, Andriy Mnih
Monte Carlo Gradient Estimation in Machine Learning. J. Mach. Learn. Res. 21 (2020)
Mihaela Rosca, Theophane Weber, Arthur Gretton, Shakir Mohamed
A case for new neural network smoothness constraints. CoRR (2020)
Mihaela Rosca, Theophane Weber, Arthur Gretton, Shakir Mohamed
A case for new neural network smoothness constraints. ICBINB@NeurIPS (2020)
Shakir Mohamed, Mihaela Rosca, Michael Figurnov, Andriy Mnih
Monte Carlo Gradient Estimation in Machine Learning. CoRR (2019)
Cyprien de Masson d'Autume, Mihaela Rosca, Jack W. Rae, Shakir Mohamed
Training language GANs from Scratch. CoRR (2019)
Yan Wu, Mihaela Rosca, Timothy P. Lillicrap
Deep Compressed Sensing. CoRR (2019)
Cyprien de Masson d'Autume, Shakir Mohamed, Mihaela Rosca, Jack W. Rae
Training Language GANs from Scratch. NeurIPS (2019)
Yan Wu, Mihaela Rosca, Timothy P. Lillicrap
Deep Compressed Sensing. ICML (2019)
Mihaela Rosca, Balaji Lakshminarayanan, Shakir Mohamed
Distribution Matching in Variational Inference. CoRR (2018)
Suman V. Ravuri, Shakir Mohamed, Mihaela Rosca, Oriol Vinyals
Learning Implicit Generative Models with the Method of Learned Moments. ICML (2018)
William Fedus, Mihaela Rosca, Balaji Lakshminarayanan, Andrew M. Dai, Shakir Mohamed, Ian J. Goodfellow
Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Step. ICLR (Poster) (2018)
Suman V. Ravuri, Shakir Mohamed, Mihaela Rosca, Oriol Vinyals
Learning Implicit Generative Models with the Method of Learned Moments. CoRR (2018)
Mihaela Rosca, Balaji Lakshminarayanan, David Warde-Farley, Shakir Mohamed
Variational Approaches for Auto-Encoding Generative Adversarial Networks. CoRR (2017)
William Fedus, Mihaela Rosca, Balaji Lakshminarayanan, Andrew M. Dai, Shakir Mohamed, Ian J. Goodfellow
Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Step. CoRR (2017)
Mihaela Rosca, Thomas M. Breuel
Sequence-to-sequence neural network models for transliteration. CoRR (2016)