Abstract

Categorical homotopy theory provides the precise mathematical framework for modeling homotopy type theory. We use structures from abstract homotopy theory, specifically model categories and -toposes1, to provide semantics for type-theoretic constructs such as Martin-Löf identity types and the univalence axiom.

Awodey-Warren Semantics

In the categorical interpretation, contexts and types are interpreted as objects and morphisms in a category equipped with a weak factorization system. Dependent types are modeled as fibrations.

Identity Type

An identity type is modeled using path objects in a model category. The reflexivity term corresponds to the weak equivalence in the factorization of the diagonal map , while the identity elimination rule (J-rule) is validated by the lifting property of fibrations against trivial cofibrations.

Simplicial and Cubical Models

We require specific combinatorial model categories to validate advanced features of homotopy type theory.

Relation to Higher Toposes

Homotopy type theory serves as the internal language of -toposes1. Just as extensional dependent type theory is the internal language of locally cartesian closed categories, extending the theory with the univalence axiom categorifies the semantics to the -categorical2 setting.

References

  • Kapulkin, K., Lumsdaine, P. L., & Pelayo, A. The simplicial model of univalent foundations.
  • Cohen, C., Coquand, T., Huber, S., & Mörtberg, A. Cubical type theory: a constructive interpretation of the univalence axiom.
  • Shulman, M. Homotopy type theory: the logic of space.
  • riehl2014-categorical-homotopy-theory

Footnotes

  1. Infinity-Topos 2

  2. Infinity Category Theory