Lectures onType Theory
ch:univalent-set-mathematics: cardinals and ordinals
appendix sectionsolutions

ch:univalent-set-mathematics: cardinals and ordinals

exercise 74.14.

On representatives, the inverse of (A+B)+CA+(B+C) sends inl(inla) to inla, inl(inrb) to inr(inlb), and inrc to inr(inrc); the inverse of the commutation map exchanges the two summands, and the inverse of 0+AA is ainra. The product maps use the same reassociation and exchange of coordinates, with a(,a) for the unit. The two distributivity inverses send inl(a,b) and inr(a,c) to (a,inlb) and (a,inrc), and similarly on the other side. Case analysis for sums and componentwise calculation for products verify both composites.

For exponentials, the inverse of (B+CA)(BA)×(CA) is copairing, the inverse of C(A×B)(CA)×(CB) pairs the two values, and the inverse of currying is evaluation on pairs. The empty- and unit-domain inverses are the unique empty function and evaluation at . Function extensionality verifies both composites. Truncation induction and univalence turn these equivalences into the cardinal equalities.

If f:AA and g:BB, the maps f+g:A+BA+B and f×g:A×BA×B are injective by case analysis and componentwise injectivity. If f:AA, then postcomposition (CA)(CA),hfh is injective by function extensionality and injectivity of f. Eliminating the representative injections into the propositional target proves compatibility with + and and base monotonicity |A||C||A||C|.

exercise 74.15.

Fix u:acc(a) and induct on u with motive P(a,u):=v:acc(a)u=v. In the constructor case write u=acc<(a,h1) and v=acc<(a,h2). For every b:A and r:b<a, the induction hypothesis applied to h1(b,r) and h2(b,r) gives h1(b,r)=h2(b,r). Function extensionality in r and then b gives h1=h2, so congruence of the accessibility constructor gives u=v. Thus acc(a) is a mere proposition. A dependent product of mere propositions is a mere proposition, so a:Aacc(a) is one as well.

exercise 74.16.

For a simulation f:AB, use well-founded induction on a with the strengthened motive P(a):=a:A(f(a)=f(a))(a=a). Given f(a)=f(a) and c<a, simulation at a gives, under a propositional truncation, c<a with f(c)=f(c). Eliminate the truncation into the proposition c=c and apply P(c). Conversely, a predecessor c<a has an image predecessor c<a, and P(c) again gives c=c. Extensionality of A now gives a=a, proving injectivity.

For simulations f,g:AB, induct on a. If b<f(a), simulation for f gives a<a with f(a)=b; the induction hypothesis changes this to g(a)=b, hence b<g(a). The converse uses g in the same way. Extensionality of B gives f(a)=g(a), and function extensionality gives f=g. If f:AB and g:BA are simulations, uniqueness applied to gf and idA, and then to fg and idB, proves that they are inverse relation isomorphisms.

exercise 210.4.

Suppose a:A and p:e(a)=d. Function congruence at a gives e(a)(a)=d(a)=not(e(a)(a)). Boolean elimination on e(a)(a) reduces this equality either to false=true or to true=false; Boolean separation refutes both cases. Thus the fiber of e over d is empty. If e were surjective, its surjectivity witness at d would give such an a and equality. The empty target is a proposition, so the truncation may be eliminated into the contradiction. The construction uses function congruence and Boolean separation, but it never decides an arbitrary proposition and therefore does not use excluded middle.

Search the book

Type to search the local edition.