10.8k citations
- Université de MontréalCA35 papers
- McGill UniversityCA24 papers
- Centre National de la Recherche ScientifiqueFR19 papers
- École de Technologie SupérieureCA13 papers
- Louisiana State UniversityUS12 papers
- Stanford UniversityUS12 papers
- Concordia UniversityCA11 papers
- Georgia Institute of TechnologyUS11 papers
- Sorbonne UniversitéFR11 papers
- Université Paris-SaclayFR11 papers
- University College LondonGB11 papers
- Columbia UniversityUS10 papers
9 papers · 2 filters
Real-Time Reinforcement Learning
Simon Ramstedt, Christopher Pal
Markov Decision Processes (MDPs), the mathematical framework underlying most algorithms in Reinforcement Learning (RL), are often used in a way that wrongfully assumes that the sta…
Training Modern Deep Neural Networks for Memory-Fault Robustness
Ghouthi Boukli Hacene, François Leduc-Primeau, Amal Ben Soussia +2
Because deep neural networks (DNNs) rely on a large number of parameters and computations, their implementation in energy-constrained systems is challenging. In this paper, we inve…
Neural Multisensory Scene Inference
Jae Hyun Lim, Pedro O. Pinheiro, Negar Rostamzadeh +2
For embodied agents to infer representations of the underlying 3D physical world they inhabit, they should efficiently combine multisensory cues from numerous trials, e.g., by look…
Scheduling optimization of parallel linear algebra algorithms using Supervised Learning
G. Laberge, S. Shirzad, P. Diehl +3
Linear algebra algorithms are used widely in a variety of domains, e.g machine learning, numerical physics and video games graphics. For all these applications, loop-level parallel…
Subspace Determination through Local Intrinsic Dimensional Decomposition: Theory and Experimentation
Ruben Becker, Imane Hafnaoui, Michael E. Houle +2
Axis-aligned subspace clustering generally entails searching through enormous numbers of subspaces (feature combinations) and evaluation of cluster quality within each subspace. In…
HyperNOMAD: Hyperparameter optimization of deep neural networks using mesh adaptive direct search
Dounia Lakhmiri, Sébastien Le Digabel, Christophe Tribes
The performance of deep neural networks is highly sensitive to the choice of the hyperparameters that define the structure of the network and the learning process. When facing a ne…