A Hardware-Software Blueprint for Flexible Deep Learning Specialization

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

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.