arXiv:1711.03016cs.PL
DLVM: A modern compiler infrastructure for deep learning systems
Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.
Abstract
Deep learning software demands reliability and performance. However, many of the existing deep learning frameworks are software libraries that act as an unsafe DSL in Python and a computation graph interpreter. We present DLVM, a design and implementation of a compiler infrastructure with a linear algebra intermediate representation, algorithmic differentiation by adjoint code generation, domain-specific optimizations and a code generator targeting GPU via LLVM. Designed as a modern compiler infrastructure inspired by LLVM, DLVM is more modular and more generic than existing deep learning compiler frameworks, and supports tensor DSLs with high expressivity. With our prototypical staged DSL embedded in Swift, we argue that the DLVM system enables a form of modular, safe and performant frameworks for deep learning.
Compiler techniques to speed-up deep learning have seen a lot of recent research activity. The article is about Intel's nGraph and provides comparisons to Google's XLA: https://www.tensorflow.org/performance/xla/ Amazon and University of Washington have NNVM…
chrisaycockView on HNParent context