Learning to Compile Programs to Neural Networks

Abstract

A neural surrogate of a program is a neural network that mimics the behavior of a program. Researchers have used these neural surrogates to automatically tune program inputs, adapt programs to new settings, and accelerate computations. Researchers traditionally develop neural surrogates by training on input-output examples from a single program. Alternatively, language models trained on a large dataset including many programs can consume program text, to act as a neural surrogate. Using a language model to both generate a surrogate and act as a surrogate, however, leading to a trade-off between resource consumption and accuracy. We present neural surrogate compilation, a technique for producing neural surrogates directly from program text without coupling neural surrogate generation and execution. We implement neural surrogate compilers using hypernetworks trained on a dataset of C programs and find that they produce neural surrogates that are 1.9-9.5 as data-efficient, produce visual results that are 1.0-1.3 more similar to ground truth, and train in 4.3-7.3 fewer epochs than neural surrogates trained from scratch.

Tasks

Reproductions