Relay: A High-Level Compiler for Deep Learning
This addresses the problem of expressivity, composability, and portability in deep learning compilation for researchers and practitioners, though it is incremental as it builds on existing IRs.
The paper tackles the challenge of extending deep learning frameworks to diverse models and hardware by presenting Relay, a high-level compiler with a functional, statically typed intermediate representation that unifies existing IRs and supports domain-specific optimizations, achieving competitive performance across CPUs, GPUs, and accelerators.
Frameworks for writing, compiling, and optimizing deep learning (DL) models have recently enabled progress in areas like computer vision and natural language processing. Extending these frameworks to accommodate the rapidly diversifying landscape of DL models and hardware platforms presents challenging tradeoffs between expressivity, composability, and portability. We present Relay, a new compiler framework for DL. Relay's functional, statically typed intermediate representation (IR) unifies and generalizes existing DL IRs to express state-of-the-art models. The introduction of Relay's expressive IR requires careful design of domain-specific optimizations, addressed via Relay's extension mechanisms. Using these extension mechanisms, Relay supports a unified compiler that can target a variety of hardware platforms. Our evaluation demonstrates Relay's competitive performance for a broad class of models and devices (CPUs, GPUs, and emerging accelerators). Relay's design demonstrates how a unified IR can provide expressivity, composability, and portability without compromising performance.