Logical Relations for Partial Features and Automatic Differentiation Correctness

Abstract

We present a simple technique for semantic, open logical relations arguments about languages with recursive types, which, as we show, follows from a principled foundation in categorical semantics. We demonstrate how it can be used to give a very straightforward proof of correctness of practical forward- and reverse-mode dual numbers style automatic differentiation (AD) on ML-family languages. The key idea is to combine it with a suitable open logical relations technique for reasoning about differentiable partial functions (a suitable lifting of the partiality monad to logical relations), which we introduce.

25 pages (18 pages + references and appendices), conference paper (the corresponding extended work can be found at arXiv:2210.07724), submitted to FoSSaCS. arXiv admin note: substantial text overlap with arXiv:2210.07724