Follow
Alena Shilova
Alena Shilova
PostDoc, Inria Lille
Verified email at inria.fr - Homepage
Title
Cited by
Cited by
Year
Training on the Edge: The why and the how
N Kukreja, A Shilova, O Beaumont, J Huckelheim, N Ferrier, P Hovland, ...
2019 IEEE International Parallel and Distributed Processing Symposium …, 2019
362019
Efficient Combination of Rematerialization and Offloading for Training DNNs
O Beaumont, L Eyraud-Dubois, A Shilova
Advances in Neural Information Processing Systems 34, 23844-23857, 2021
352021
Optimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory
J Herrmann, O Beaumont, L Eyraud-Dubois, J Hermann, A Joly, A Shilova
arXiv preprint arXiv:1911.13214, 2019
252019
Optimal memory-aware backpropagation of deep join networks
O Beaumont, J Herrmann, G Pallez, A Shilova
Philosophical Transactions of the Royal Society A 378 (2166), 20190049, 2020
202020
Optimal GPU-CPU offloading strategies for deep neural network training
O Beaumont, L Eyraud-Dubois, A Shilova
European Conference on Parallel Processing, 151-166, 2020
172020
Survey on Efficient Training of Large Neural Networks
J Gusak, D Cherniuk, A Shilova, A Katrutsa, D Bershatsky, X Zhao, ...
IJCAI-ECAI, 2022
15*2022
Pipelined model parallelism: Complexity results and memory considerations
O Beaumont, L Eyraud-Dubois, A Shilova
European Conference on Parallel Processing, 183-198, 2021
82021
A makespan lower bound for the scheduling of the tiled cholesky factorization based on alap schedule
O Beaumont, J Langou, W Quach, A Shilova
EuroPar 2020-26th International European Conference on Parallel and …, 2020
62020
Revisiting Continuous-Time Reinforcement Learning. A Study of HJB Solvers Based on PINNs and FEMs
A Shilova, T Delliaux, P Preux, B Raffin
Sixteenth European Workshop on Reinforcement Learning, 2023
22023
MadPipe: Memory Aware Dynamic Programming Algorithm for Pipelined Model Parallelism
O Beaumont, L Eyraud-Dubois, A Shilova
2022 IEEE International Parallel and Distributed Processing Symposium …, 2022
12022
Weight Offloading Strategies for Training Large DNN Models
O Beaumont, L Eyraud-Dubois, A Shilova, X Zhao
12022
An Integer Linear Programming Approach for Pipelined Model Parallelism
O Beaumont, L Eyraud-Dubois, A Shilova
Inria, 2022
12022
A makespan lower bound for the tiled cholesky factorization based on alap schedule
O Beaumont, J Langou, W Quach, A Shilova
European Conference on Parallel Processing, 134-150, 2020
12020
Learning HJB Viscosity Solutions with PINNs for Continuous-Time Reinforcement Learning
A Shilova, T Delliaux, P Preux, B Raffin
Inria Lille-Nord Europe, CRIStAL-Centre de Recherche en Informatique, Signal …, 2024
2024
Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory
O Beaumont, L Eyraud-Dubois, J Herrmann, A Joly, A Shilova
ACM Transactions on Mathematical Software, 2024
2024
AdaStop: sequential testing for efficient and reliable comparisons of Deep RL Agents
T Mathieu, R Della Vecchia, A Shilova, MC de Medeiros, H Kohler, ...
arXiv preprint arXiv:2306.10882, 2023
2023
Entropy Regularized Reinforcement Learning with Cascading Networks
R Della Vecchia, A Shilova, P Preux, R Akrour
arXiv preprint arXiv:2210.08503, 2022
2022
Memory Saving Strategies for Deep Neural Network Training
A Shilova
Université de Bordeaux, 2021
2021
Memory Efficient Deep Neural Network Training
A Shilova
European Conference on Parallel Processing, 515-519, 2021
2021
The system can't perform the operation now. Try again later.
Articles 1–19