The Sup Connective in IMALL: A Categorical Semantics
Alejandro Díaz-Caro, Octavio Malherbe
TL;DR
This work develops a categorical semantics for an IMALL-inspired language extended with the sup connective, capturing probabilistic and additive behavior via a weighted codiagonal in categories with biproducts. It shows that any symmetric monoidal closed category with biproducts and a monomorphism from a scalar semiring into End(I) suffices to interpret the language, providing both soundness and adequacy results. By internalizing probabilistic choice and vector/matrix structures through sums and scalar products in the syntax, the paper unifies probabilistic reasoning with linear logic connectives and offers a groundwork for further links to quantum computing and probabilistic programming. The categorical construction, including the map scalars, weighted codiagonal, and the delta map, provides a flexible, abstract model that extends prior PCF^R-like approaches while enabling richer connectives. This framework paves the way for applying IMALL+sup to verifiable quantum algorithms and probabilistic languages, with clear semantics grounded in biproducts and monoidal structure.
Abstract
We explore a proof language for intuitionistic multiplicative additive linear logic, incorporating the sup connective that introduces additive pairs with a probabilistic elimination, and sum and scalar products within the proof-terms. We provide an abstract characterisation of the language, revealing that any symmetric monoidal closed category with biproducts and a monomorphism from the semiring of scalars to the semiring Hom(I,I) is suitable for the job. Leveraging the binary biproducts, we define a weighted codiagonal map at the heart of the sup connective.
