N
Nicolas Vasilache
Researcher at Facebook
Publications - 51
Citations - 3121
Nicolas Vasilache is an academic researcher from Facebook. The author has contributed to research in topics: Code generation & Compiler. The author has an hindex of 25, co-authored 46 publications receiving 2770 citations. Previous affiliations of Nicolas Vasilache include University of Paris-Sud & IBM.
Papers
More filters
Posted Content
Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
Nicolas Vasilache,Oleksandr Zinenko,Theodoros Theodoridis,Priya Goyal,Zachary DeVito,William S. Moses,Sven Verdoolaege,Andrew Adams,Albert Cohen +8 more
TL;DR: A language close to the mathematics of deep learning called Tensor Comprehensions offering both imperative and declarative styles, a polyhedral Just-In-Time compiler to convert a mathematical description of a deep learning DAG into a CUDA kernel with delegated memory management and synchronization, and a compilation cache populated by an autotuner are contributed.
Proceedings Article
Fast Convolutional Nets With fbfft: A GPU Performance Evaluation
TL;DR: This work examines the performance profile of Convolutional Neural Network training on the current generation of NVIDIA Graphics Processing Units, and introduces two new Fast Fourier Transform convolution implementations: one based on NVIDIA's cuFFT library, and another based on a Facebook authored FFT implementation, fbfft, that provides significant speedups over cuFFt.
Journal ArticleDOI
Semi-automatic composition of loop transformations for deep parallelism and memory hierarchies
Sylvain Girbal,Nicolas Vasilache,Cédric Bastoul,Albert Cohen,David Parello,Marc Sigler,Olivier Temam +6 more
TL;DR: This work leverages on algorithmic advances in polyhedral code generation and has been implemented in a modern research compiler, using a semi-automatic optimization approach to demonstrate that current compilers suffer from unnecessary constraints and intricacies that can be avoided in a semantically richer transformation framework.
Posted Content
Learning Visual Features from Large Weakly Supervised Data
TL;DR: This paper trains convolutional networks on a dataset of 100 million Flickr photos and comments, and shows that these networks produce features that perform well in a range of vision problems.
Book ChapterDOI
Learning Visual Features from Large Weakly Supervised Data
TL;DR: In this paper, the authors explore the potential of leveraging massive, weakly-labeled image collections for learning good visual features, and train convolutional networks on a dataset of 100 million Flickr photos and comments.