Relay: A New IR for Machine Learning Frameworks
This work addresses the problem of balancing constraints in machine learning frameworks for industry applications, but it appears incremental as it builds on existing compiler frameworks without demonstrating new performance gains.
The authors tackled the challenge of making machine learning models efficient, expressive, and portable across hardware by proposing Relay, a new high-level intermediate representation designed as a purely-functional, statically-typed language, with a prototype integrated into the NNVM compiler framework used by Amazon's MxNet.
Machine learning powers diverse services in industry including search, translation, recommendation systems, and security. The scale and importance of these models require that they be efficient, expressive, and portable across an array of heterogeneous hardware devices. These constraints are often at odds; in order to better accommodate them we propose a new high-level intermediate representation (IR) called Relay. Relay is being designed as a purely-functional, statically-typed language with the goal of balancing efficient compilation, expressiveness, and portability. We discuss the goals of Relay and highlight its important design constraints. Our prototype is part of the open source NNVM compiler framework, which powers Amazon's deep learning framework MxNet.