Léopold Cambier

Léopold Cambier

Auto-generated profile, based on publicly available information

Léopold Cambier

Institute for Computational and Mathematical Engineering Stanford University

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2020 in SIAM Journal on Matrix Analysis and Applications
Verified
Journal Article
Closed access
LCLéopold Cambier
Léopold Cambier
CCChao Chen
Chao Chen
EBErik G. Boman
Erik G. Boman
SRSivasankaran Rajamanickam
Sivasankaran Rajamanickam
RTRaymond S. Tuminaro
Raymond S. Tuminaro
EDEric Darve
Eric Darve
We propose a new algorithm for the fast solution of large, sparse, symmetric positive-definite linear systems, spaND (sparsified Nested Dissection). It is based on nested dissection, sparsification...

1 November 2019 in Journal of Computational Physics
Verified
Journal Article
Closed access
CCChao Chen
Chao Chen
LCLeopold Cambier
Leopold Cambier
EBErik G. Boman
Erik G. Boman
SRSivasankaran Rajamanickam
Sivasankaran Rajamanickam
RTRaymond S. Tuminaro
Raymond S. Tuminaro
EDEric Darve
Eric Darve

1 January 2019 in SIAM Journal on Scientific Computing
Verified
Journal Article
Closed access
LCLéopold Cambier
Léopold Cambier
EDEric Darve
Eric Darve
Integral equations are commonly encountered when solving complex physical problems. Their discretization leads to a dense kernel matrix that is block or hierarchically low-rank. This paper proposes a new way to build a low-rank factorization of those low-rank blocks at a nearly optimal cost of $\mathcal{O}\!\left(nr\right)$ for an $n\times n$ block submatrix of rank $r$. This is done by first sampling the kernel function at new interpolation points, then selecting a subset of those using a CUR decomposition and finally using this reduced set of points as pivots for a rank-revealing LU-type factorization. We also explain how this implicitly builds an optimal interpolation basis for the kernel under consideration. We show the asymptotic convergence of the algorithm, explain its stability, and demonstrate on numerical examples that it performs very well in practice, allowing us to obtain rank nearly equal to the optimal rank at a fraction of the cost of the naive algorithm.

12 August 2022 in International Journal for Numerical Methods in Engineering
Verified
Journal Article
Closed access
BKBazyli Klockiewicz
Bazyli Klockiewicz
LCLéopold Cambier
Léopold Cambier
RHRyan Humble
Ryan Humble
HTHamdi Tchelepi
Hamdi Tchelepi
EDEric Darve
Eric Darve

16 January 2020 in arXiv
Verified
Preprint
Open access
PDF file
LCLéopold Cambier
Léopold Cambier
ABAnahita Bhiwandiwalla
Anahita Bhiwandiwalla
TGTing Gong
Ting Gong
MNMehran Nekuii
Mehran Nekuii
OEOguz H Elibol
Oguz H Elibol
HTHanlin Tang
Hanlin Tang
Training with larger number of parameters while keeping fast iterations is an increasingly adopted strategy and trend for developing better performing Deep Neural Network (DNN) models. This necessitates increased memory footprint and computational requirements for training. Here we introduce a novel methodology for training deep neural networks using 8-bit floating point (FP8) numbers. Reduced bit precision allows for a larger effective memory and increased computational speed. We name this method Shifted and Squeezed FP8 (S2FP8). We show that, unlike previous 8-bit precision training methods, the proposed method works out-of-the-box for representative models: ResNet-50, Transformer and NCF. The method can maintain model accuracy without requiring fine-tuning loss scaling parameters or keeping certain layers in single precision. We introduce two learnable statistics of the DNN tensors - shifted and squeezed factors that are used to optimally adjust the range of the tensors in 8-bits, thus minimizing the loss in information due to quantization.

6 May 2019 in arXiv
Verified
Preprint
Open access
PDF file
LCLéopold Cambier
Léopold Cambier
EDEric Darve
Eric Darve
Integral equations are commonly encountered when solving complex physical problems. Their discretization leads to a dense kernel matrix that is block or hierarchically low-rank. This paper proposes a new way to build a low-rank factorization of those low-rank blocks at a nearly optimal cost of $\mathcal{O}(nr)$ for a $n \times n$ block submatrix of rank r. This is done by first sampling the kernel function at new interpolation points, then selecting a subset of those using a CUR decomposition and finally using this reduced set of points as pivots for a RRLU-type factorization. We also explain how this implicitly builds an optimal interpolation basis for the Kernel under consideration. We show the asymptotic convergence of the algorithm, explain his stability and demonstrate on numerical examples that it performs very well in practice, allowing to obtain rank nearly equal to the optimal rank at a fraction of the cost of the naive algorithm.