A Hardware–Software Blueprint for Flexible Deep Learning Specialization

Thierry Moreau,Tianqi Chen,Luis Vega,Jared Roesch,Lianmin Zheng,Josh Fromm,Ziheng Jiang,Luis Ceze,Carlos Guestrin,Arvind Krishnamurthy

arXiv: Learning（2019）

Cited 177|Views700

No score

Abstract

This article describes the Versatile Tensor Accelerator (VTA), a programmable DL architecture designed to be extensible in the face of evolving workloads. VTA achieves “flexible specialization” via a parameterizable architecture, two-level Instruction Set Architecture (ISA), and a Just in Time (JIT) compiler.

Translated text

Key words

Machine learning,Computer architecture,Runtime,Task analysis,Computational modeling,Deep learning

AI Read Science

Must-Reading Tree

Example

Generate MRT to find the research sequence of this paper

Chat Paper

Summary is being generated by the instructions you defined