Logical Relations for Partial Features and Automatic Differentiation Correctness

10/16/2022
by   Fernando Lucatelli Nunes, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/14/2022

Automatic Differentiation for ML-family languages: correctness via logical relations

We give a simple, direct and reusable logical relations technique for la...
research
02/19/2020

On the Versatility of Open Logical Relations: Continuity, Automatic Differentiation, and a Containment Theorem

Logical relations are one of the most powerful techniques in the theory ...
research
10/01/2021

CHAD for Expressive Total Languages

We show how to apply forward and reverse mode Combinatory Homomorphic Au...
research
07/10/2020

Denotational Correctness of Foward-Mode Automatic Differentiation for Iteration and Recursion

We present semantic correctness proofs of forward-mode Automatic Differe...
research
01/07/2020

Correctness of Automatic Differentiation via Diffeologies and Categorical Gluing

We present semantic correctness proofs of Automatic Differentiation (AD)...
research
01/31/2023

On the Correctness of Automatic Differentiation for Neural Networks with Machine-Representable Parameters

Recent work has shown that automatic differentiation over the reals is a...
research
07/25/2019

An Introduction to Logical Relations

Logical relations (LR) have been around for many years, and today they a...

Please sign up or login with your details

Forgot password? Click here to reset