Eric Petit

  • Intel
  • Versailles Saint-Quentin-en-Yvelines University, France

According to our database1, Eric Petit authored at least 28 papers between 2012 and 2023.

Collaborative distances:



In proceedings 
PhD thesis 


Online presence:



Stochastic Rounding Variance and Probabilistic Bounds: A New Approach.
SIAM J. Sci. Comput., October, 2023

Bounds on non-linear errors for variance computation with stochastic rounding.
CoRR, 2023

Lasa: Abstraction and Specialization for Productive and Performant Linear Algebra on FPGAs.
Proceedings of the 31st IEEE Annual International Symposium on Field-Programmable Custom Computing Machines, 2023

A BF16 FMA is All You Need for DNN Training.
IEEE Trans. Emerg. Top. Comput., 2022

FASE: A Fast, Accurate and Seamless Emulator for Custom Numerical Formats.
Proceedings of the Machine Learning and Knowledge Discovery in Databases, 2022

The Positive Effects of Stochastic Rounding in Numerical Algorithms.
Proceedings of the 29th IEEE Symposium on Computer Arithmetic, 2022

Confidence Intervals for Stochastic Arithmetic.
ACM Trans. Math. Softw., 2021

A Study of the Effects and Benefits of Custom-Precision Mathematical Libraries for HPC Codes.
IEEE Trans. Emerg. Top. Comput., 2021

Dynamically Adapting Floating-Point Precision to Accelerate Deep Neural Network Training.
Proceedings of the 20th IEEE International Conference on Machine Learning and Applications, 2021

Shadow computation with BFloat16 to estimate the numerical accuracy of summations.
Proceedings of the 28th IEEE Symposium on Computer Arithmetic, 2021

Comparing perturbation models for evaluating stability of neuroimaging pipelines.
Int. J. High Perform. Comput. Appl., 2020

Evaluating Mixed-Precision Arithmetic for 3D Generative Adversarial Networks to Simulate High Energy Physics Detectors.
Proceedings of the 19th IEEE International Conference on Machine Learning and Applications, 2020

Custom-Precision Mathematical Library Explorations for Code Profiling and Optimization.
Proceedings of the 27th IEEE Symposium on Computer Arithmetic, 2020

Comparing Perturbation Models for Evaluating Stability of Post-Processing Pipelines in Neuroimaging.
CoRR, 2019

Scalable Fast Multipole Method for Electromagnetic Simulations.
Proceedings of the Computational Science - ICCS 2019, 2019

Automatic Exploration of Reduced Floating-Point Representations in Iterative Methods.
Proceedings of the Euro-Par 2019: Parallel Processing, 2019

Asynchronous and multithreaded communications on irregular applications using vectorized divide and conquer approach.
J. Parallel Distributed Comput., 2018

Scalable Work-Stealing Load-Balancer for HPC Distributed Memory Systems.
Proceedings of the Euro-Par 2018: Parallel Processing Workshops, 2018

VeriTracer: Context-enriched tracer for floating-point arithmetic analysis.
Proceedings of the 25th IEEE Symposium on Computer Arithmetic, 2018

A software scheduling solution to avoid corrupted units on GPUs.
J. Parallel Distributed Comput., 2016

Verificarlo: Checking Floating Point Accuracy through Monte Carlo Arithmetic.
Proceedings of the 23nd IEEE Symposium on Computer Arithmetic, 2016

CERE: LLVM-Based Codelet Extractor and REplayer for Piecewise Benchmarking and Optimization.
ACM Trans. Archit. Code Optim., 2015

Task-Based Parallelization of Unstructured Meshes Assembly Using D&C Strategy.
Proceedings of the 2014 IEEE International Conference on High Performance Computing and Communications, 2014

Adaptive sampling for performance characterization of application kernels.
Concurr. Comput. Pract. Exp., 2013

GPUburn: A system to test and mitigate GPU hardware failures.
Proceedings of the 2013 International Conference on Embedded Computer Systems: Architectures, 2013

Divide and Conquer Parallelization of Finite Element Method Assembly.
Proceedings of the Parallel Computing: Accelerating Computational Science and Engineering (CSE), 2013

Binary Instrumentation for Scalable Performance Measurement of OpenMP Applications.
Proceedings of the Parallel Computing: Accelerating Computational Science and Engineering (CSE), 2013

ASK: Adaptive Sampling Kit for Performance Characterization.
Proceedings of the Euro-Par 2012 Parallel Processing - 18th International Conference, 2012