activity
20192022
most citedOptimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory

13 citations · 22 across the 5 of their papers we have counts for

collaborators

6 papers

cs.LG20228 cited

Survey on Large Scale Neural Network Training

Julia Gusak, Daria Cherniuk, Alena Shilova +8

Modern Deep Neural Networks (DNNs) require significant memory to store weight, activations, and other intermediate tensors during training. Hence, many models do not fit one GPU de…

cs.DC2022

I/O-Optimal Algorithms for Symmetric Linear Algebra Kernels

Olivier Beaumont, Lionel Eyraud-Dubois, Mathieu Vérité +1

In this paper, we consider two fundamental symmetric kernels in linear algebra: the Cholesky factorization and the symmetric rank- update (SYRK), with the classical three nested…

cs.AI2020

Geometric Deep Reinforcement Learning for Dynamic DAG Scheduling

Nathan Grinsztajn, Olivier Beaumont, Emmanuel Jeannot +1

In practice, it is quite common to face combinatorial optimization problems which contain uncertainty along with non-determinism and dynamicity. These three properties call for app…

cs.LG201913 cited

Optimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory

Julien Herrmann, Olivier Beaumont, Lionel Eyraud-Dubois +3

This paper introduces a new activation checkpointing method which allows to significantly decrease memory usage when training Deep Neural Networks with the back-propagation algorit…

cs.DC2019

Scheduling on Two Types of Resources: a Survey

Olivier Beaumont, Louis-claude Canon, Lionel Eyraud-Dubois +5

The evolution in the design of modern parallel platforms leads to revisit the scheduling jobs on distributed heterogeneous resources. The goal of this survey is to present the main…

cs.DC20191 cited

Training on the Edge: The why and the how

Navjot Kukreja, Alena Shilova, Olivier Beaumont +4

Edge computing is the natural progression from Cloud computing, where, instead of collecting all data and processing it centrally, like in a cloud computing environment, we distrib…