Homotopy Type Theory homotopy type theory > history (Rev #10, changes)

Showing changes from revision #9 to #10: Added | Removed | Changed

\tableofcontents

\section{Idea}

Homotopy type theory is a framework of dependent type theories which additionally consists of

  • identity typesidentity types

  • dependent product types?dependent product types

  • dependent sum types?dependent sum types

  • univalent universesunivalent universes

  • inductive types?inductive types, higher inductive types, inductive type families, et cetera., higher inductive types, inductive type families?, et cetera.

\section{Presentation}

The model of homotopy type theory we shall be presenting here is the objective type theory version of homotopy type theory. There are multiple reasons for this:

  • Since objective type theory lacks definitional equality,

    • The ruleset is simpler in the objective type theory model of homotopy type theory than other models such as Martin-Löf type theory, cubical type theory, or higher observational type theory

    • The results in objective type theory are more general than in models which use definitional equality

    • It is similar to other non-type theory foundations such as the various flavors of set theory, since it also only has one notion of equality, which is propositional equality in both objective type theory and set theory, and uses propositional equality to define terms and types.

  • From a more practical standpoint, objective type theory not only has decidable type checking, it has polynomial (quadratic) time type checking, which makes it efficient from a computational standpoint.

\subsection{Types In a similar manner, for simplicity and terms} ease of presentation, we shall also follow the HoTT book in not including a separateAtypeA \; \mathrm{type} judgment and rather stipulating that every type is an element of a Russell universe.

As \subsection{Judgments with and any contexts} type theory, the basic judgments of this model are judgingAA to be a type, AtypeA \; \mathrm{type}, and judging aa to be a term of AA, a:Aa:A, and judging Γ\Gamma to be a context, Γctx\Gamma \; \mathrm{ctx}. Contexts are lists of type and term judgments.

\subsection{Dependent We types introduce and two sections} judgments in the model: typing judgments, where we judgeaa to be an element of AA, a:Aa:A, and context judgments, where we judge Γ\Gamma to be a context, Γctx\Gamma \; \mathrm{ctx}. Contexts are lists of typing judgments a:Aa:A, b:Bb:B, c:Cc:C, et cetera, and are formalized by the rules for the empty context and extending the context by a typing judgment

A dependent type is a type BB in the context of the variable judgment x:Ax:A, x:ABtypex:A \vdash B \; \mathrm{type}. The dependent type is usually written as B(x)B(x) to indicate the dependence upon xx.

()ctxΓctxΓA:U i(Γ,a:A)ctx\frac{}{() \; \mathrm{ctx}} \qquad \frac{\Gamma \; \mathrm{ctx} \quad \Gamma \vdash A:\mathrm{U}_i}{(\Gamma, a:A) \; \mathrm{ctx}}

A \subsection{Structural section rules} is a termb:Bb:B in the context of the variable judgment x:Ax:A, x:Ab:Bx:A \vdash b:B. Similarly for the case for dependent types, sections are usually written as b(x)b(x) to indicate its dependence upon xx.

\subsection{Variable, The Substitution, variable and rule Weakening states rules} that we may derive a typing judgment if the typing judgment is in the context already:

Γ,a:A,ΔctxΓ,a:A,Δa:A\frac{\vdash \Gamma, a:A, \Delta \mathrm{ctx}}{\vdash \Gamma, a:A, \Delta \vdash a:A}

Let 𝒥\mathcal{J} be any arbitrary judgment. Then we have the following rules:

The weakening rule:

Γ,Δ𝒥ΓA:𝒰 iΓ,a:A,Δ𝒥\frac{\Gamma, \Delta \vdash \mathcal{J} \quad \Gamma \vdash A:\mathcal{U}_i}{\Gamma, a:A, \Delta \vdash \mathcal{J}}

The substitution rule:

Γa:AΓ,b:A,Δ𝒥Γ,Δ[a/b]𝒥[a/b]\frac{\Gamma \vdash a:A \quad \Gamma, b:A, \Delta \vdash \mathcal{J}}{\Gamma, \Delta[a/b] \vdash \mathcal{J}[a/b]}

The weakening and substitution rules are admissible rules: they do not need to be explicitly included in the type theory as they could be proven by induction on the structure of all possible derivations.

\subsection{Russell universes}

Universes are the types of types in type theory.

We introduce a universe hierarchy of universesUU i U U_i , with indexed rules by for natural universe numbers formation and type reflectionii in the metatheory, with rules for universe formation and cumulativity.

ΓctxΓUU i:typeU i+1ΓA:UU iΓA:typeU i+1 \frac{\Gamma \; \mathrm{ctx}}{\Gamma \vdash U U_i \; : \mathrm{type}} U_{i + 1}} \qquad \frac{\Gamma \vdash A:U}{\Gamma A:U_i}{\Gamma \vdash A A:U_{i \; + \mathrm{type}} 1}}

\subsection{Sections and dependent types}

A section is a term b:Bb:B in the context of the variable judgment x:Ax:A, x:Ab:Bx:A \vdash b:B. Sections are usually written as b(x)b(x) to indicate its dependence upon xx.

A dependent type is a section of a universe B:U iB:U_i in the context of the variable judgment x:Ax:A, x:AB:U ix:A \vdash B:U_i, and since they are sections, they are usually written as B(x)B(x).

\subsection{Equality}

Equality in type theory is represented by the identity type, which is also called the path type or identification type. The terms of the identity type could be called paths or identifications.

Equality comes with a formation rule, an introduction rule, an elimination rule, and a computation rule:

Formation rule for identity types:

ΓA:U iΓ,a:A,b:Aa= Ab:U i\frac{\Gamma \vdash A:U_i}{\Gamma, a:A, b:A \vdash a =_A b:U_i}

Introduction rule for identity types:

ΓA:U iΓ,a:Arefl A(a):a= Aa\frac{\Gamma \vdash A:U_i}{\Gamma, a:A \vdash \mathrm{refl}_A(a) : a =_A a}

Elimination rule for identity types:

Γ,a:A,b:A,p:a= Ab,Δ(a,b,p)C(a,b,p):U iΓ,a:A,Δ(a,a,refl A(a))t:C(a,a,refl A(a))Γ,a:A,b:A,p:a= Ab,Δ(a,b,p)J(t,a,b,p):C(a,b,p)\frac{\Gamma, a:A, b:A, p:a =_A b, \Delta(a, b, p) \vdash C(a, b, p):U_i \quad \Gamma, a:A, \Delta(a, a, \mathrm{refl}_A(a)) \vdash t:C(a, a, \mathrm{refl}_A(a))}{\Gamma, a:A, b:A, p:a =_A b, \Delta(a, b, p) \vdash J(t, a, b, p):C(a, b, p)}

Conversion rules for identity types:

Γ,a:A,b:A,p:a= Ab,Δ(a,b,p)C(a,b,p):U iΓ,a:A,Δ(a,a,refl A(a))t:C(a,a,refl A(a))Γ,a:A,b:A,p:a= Ab,Δ(a,b,p)β = A(a):J(t,a,a,refl(a))= C(a,a,refl A(a))t\frac{\Gamma, a:A, b:A, p:a =_A b, \Delta(a, b, p) \vdash C(a, b, p):U_i \quad \Gamma, a:A, \Delta(a, a, \mathrm{refl}_A(a)) \vdash t:C(a, a, \mathrm{refl}_A(a))}{\Gamma, a:A, b:A, p:a =_A b, \Delta(a, b, p) \vdash \beta_{=_A}(a) : J(t, a, a, \mathrm{refl}(a)) =_{C(a, a, \mathrm{refl}_A(a))} t}

\subsection{Definitions}

In mathematics, many times one would want to define an element to be another element of a type. In order to define an element a:Aa:A to be an element b:Ab:A, one says that element a:Aa:A comes with an identification def(a,b):a= Ab\mathrm{def}(a, b):a =_A b. Since types are elements of universes, one defines a type A:𝒰 iA:\mathcal{U}_i to be a type B:𝒰 iB:\mathcal{U}_i if the type A:𝒰 iA:\mathcal{U}_i comes with an identification def(A,B):A= 𝒰 iB\mathrm{def}(A, B):A =_{\mathcal{U}_i} B.

Sometimes, for ease of simplicity, the identity type is simply written a=ba = b for elements a:Aa:A and b:Ab:A. The type argument AA for the identity type then becomes an implicit argument. Thus, in order to define aa to be bb, one says that aa comes with an identification def(a,b):a=b\mathrm{def}(a, b):a = b.

In a proof assistant or some other program, elaboration is needed to expand out all the implicit arguments to get the right type a= Aba =_A b.

\subsection{Functions}

\subsection{Dependent products}

\subsection{Products}

\subsection{Dependent sums}

\subsection{Sums}

\subsection{Empty type}

\subsection{Unit type}

\subsection{Booleans}

\subsection{Interval}

\subsection{Function extensionality}

\subsection{Equivalences}

\subsection{Univalence}

\subsection{Natural numbers}

\section{References}

  • Homotopy Type Theory: Univalent Foundations of Mathematics, The Univalent Foundations Program, Institute for Advanced Study, 2013. (web, pdf)

  • Egbert RijkeEgbert Rijke, , Introduction to Homotopy Type Theory, Cambridge Studies in Advanced Mathematics, Cambridge University Press (pdf) (478 pages)

  • Benno van den BergBenno van den Berg, Martijn den Besten, , Martijn den Besten?, Quadratic type checking for objective type theory (arXiv:2102.00905)

\section{External links}

Revision on September 30, 2022 at 17:53:17 by Anonymous?. See the history of this page for a list of all contributions to it.