Latte: a language, compiler, and runtime for elegant and efficient deep neural networks
| Autor: | Rajkishore Barik, Tatiana Shpeisman, Ehsan Totoni, Chick Markley, Armando Fox, Hai Liu, Leonard Truong |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Domain-specific language
Speedup Computer science Parallel computing 02 engineering and technology 010501 environmental sciences computer.software_genre 01 natural sciences 0202 electrical engineering electronic engineering information engineering Implementation 0105 earth and related environmental sciences Xeon business.industry Programming language Deep learning 020207 software engineering Computer Graphics and Computer-Aided Design x86 Distributed memory 020201 artificial intelligence & image processing Artificial intelligence Compiler business Machine code computer Software |
| Zdroj: | PLDI |
| ISSN: | 1558-1160 0362-1340 |
| DOI: | 10.1145/2980983.2908105 |
| Popis: | Deep neural networks (DNNs) have undergone a surge in popularity with consistent advances in the state of the art for tasks including image recognition, natural language processing, and speech recognition. The computationally expensive nature of these networks has led to the proliferation of implementations that sacrifice abstraction for high performance. In this paper, we present Latte, a domain-specific language for DNNs that provides a natural abstraction for specifying new layers without sacrificing performance. Users of Latte express DNNs as ensembles of neurons with connections between them. The Latte compiler synthesizes a program based on the user specification, applies a suite of domain-specific and general optimizations, and emits efficient machine code for heterogeneous architectures. Latte also includes a communication runtime for distributed memory data-parallelism. Using networks described using Latte, we demonstrate 3-6x speedup over Caffe (C++/MKL) on the three state-of-the-art ImageNet models executing on an Intel Xeon E5-2699 v3 x86 CPU. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|