are left and right inverses of b, respectively, then: (b b 1 and b 1 = b 1 b 1 id T = b 1 b) b 1 so they are the same! r ) = (b 1 r = id S b 1 r = b 1

Transcription

1 Lecture 1. The Category of Sets PCMI Summer 2015 Undergraduate Lectures on Fag Varieties Lecture 1. Some basic set theory, a moment of categorica zen, and some facts about the permutation groups on n etters. A set is a coection of eements. The standard finite sets are: [n] := {1, 2, 3,..., n} i.e. the coections of the first n natura numbers. Remark. The empty set is the unique set with zero eements. Notation. S is the number of eements (cardinaity) of a finite set S. A map: f : S T is a rue for assigning a unique eement t T to each eement s S. This is written in function notation as: f(s) = t. The map f is injective, surjective or bijective, respectivey, if: (inj) Each t T is assigned to at most one s S. (surj) Each t T is assigned to at east one s S. (bij) Each t T is assigned to exacty one s S. Maps can be composed. If f : S T and g : T U, then: (g f)(s) = g(f(s)) defines the composition g f : S U, and if h : U V, then: h (g f) = (h g) f In other words, composition of maps is an associative operation. Every set S comes equipped with the identity sef-map: id S : S S id S (s) = s Evidenty, f id S = f and id T f = f for a maps f : S T. Each bijective map b : S T has a two-sided inverse b 1 : T S, i.e. b 1 b = id S and b b 1 = id T Remark. The fact that the eft and right inverses of a bijection are the same is a consequence of associativity and the properties of the identity. If b 1 and b 1 r b 1 = b 1 id T = b 1 so they are the same! are eft and right inverses of b, respectivey, then: (b b 1 r ) = (b 1 b) b 1 r = id S b 1 r = b 1 r 1

2 2 Moment of Zen. The category of sets consists of the two coections: (a) The coection of a sets, (b) The coection of a maps of sets which we visuaize as a universe of points (sets) and arrows (maps). More precisey: Definition 1.1. A category C consists of two coections: (a) The coection ob(c) of objects X of C, and (b) The coection mor(c) of morphisms f : X Y between objects equipped with a composition aw with the foowing properties: (i) The composition aw is associative (in the sense we discussed). (ii) Each object has an identity morphism id X : X X such that f id X = f = id Y f for a objects X, Y and morphisms f : X Y (iii) By the argument above, inverses (when they exist) are two-sided. This is designed so that sets and maps form the category Sets. The morphisms from an object to itsef are caed endomorphisms and the morphisms with (two-sided) inverses are caed isomorphisms. An automorphism is an endomorphism that is aso an isomorphism. Let: End(S) and Aut(S) be the sets of endomorphisms and automorphisms of a set S. Tupe Notation. Each map f : [n] S is the n-tupe of its vaues: (f(1), f(2),..., f(n)) S n In particuar, an n-tupe of eements of [n] is an eement f End([n]), and if the eements are distinct, then f Aut([n]). We concude that End([n]) = n n and Aut([n]) = n! Since every pair of endomorphisms can be composed, we can form a composition tabe for the endomorphisms of a finite set: Exampe 1.1. The eements of End([2]) are (1, 1), (1, 2), (2, 1), (2, 2) with composition tabe: g f g = (1, 1) (1, 2) (2, 1) (2, 2) f = (1, 1) (1, 1) (1, 1) (2, 2) (2, 2) (1, 2) (1, 1) (1, 2) (2, 1) (2, 2) (2, 1) (1, 1) (2, 1) (1, 2) (2, 2) (2, 2) (1, 1) (2, 2) (1, 1) (2, 2)

3 3 Notice that aready in this case, composition is not commutative! Our first interesting exampe of a representation is the foowing: Definition 1.2: Let f End([n]). Then: sgn(f) = f(j) f(i) pairs i<j is the characteristic sign function of the endomorphism. Remark. We coud have aso chosen i > j in a pair, since: f(j) f(i) = f(i) f(j) i j Proposition 1.1. (a) sgn(f) = 0 if and ony if f Aut([n]). (b) Otherwise sgn(f) = ±1. (c) sgn is a mutipicative function, i.e. sgn(f g) = sgn(f) sgn(g) for a pairs of endomorphisms f, g. Proof. Since [n] is finite, f fais to be an automorphism if and ony if it fais to be injective, and f fais to be injective if and ony if the numerator of some factor of sgn(f) is zero. This is (a). For (b), et f Aut([n]). Then the pairs {f(i), f(j)} vary over a two-eement subsets of [n] as the pairs {i, j} vary over a two-eement subsets. It foows that i<j f(j) f(i) = i<j and therefore that i<j (f(j) f(i)) = ± i<j (), which gives (b). Notice that it may be the case that i < j but f(i) > f(j). In fact, the number of such crossings determines whether sgn(f) is +1 or 1. Let h = f g. If f or g is not an automorphism, then h is not, and: sgn(h) = 0 = sgn(f) sgn(g) Otherwise, g in particuar is an automorphism, and: sgn(h) = i<j f(g(j)) f(g(i)) = i<j f(g(j)) f(g(i)) g(j) g(i) g(j) g(i) The product of the second factors gives sgn(g), and the product of the first factors (and the remark above) gives sgn(f). Remark. The inverse of a composition of automorphisms satisfies: (f g) 1 = g 1 f 1

4 4 Notation. Sef-compositions of an automorphism are written as powers: f 2 = f f, f 3 = f f f, etc Definition 1.3. Aut([n]) is caed the permutation group Perm(n). Exampe 1.2. Perm(3) consists of six eements: id = (1, 2, 3), (1, 3, 2), (2, 1, 3), (2, 3, 1), (3, 1, 2), (3, 2, 1) These eements fa into three distinct casses: (id) id = (1, 2, 3) fixes each eement. (tr) (1, 3, 2), (2, 1, 3) and (3, 2, 1) each transpose two eements. (cyc) (2, 3, 1), (3, 1, 2) each cyce through a the eements. Permutations have an extremey usefu cyce notation: Definition 1.4. Let s S and f Aut(S). The sequence: s, f(s), f 2 (s), f 3 (s),... of eements of S is the orbit of s under the automorphism f. Proposition 1.2. The orbit of each i [n] cyces under a permutation σ Perm(n), i.e. there is a distinct sequence i 1,..., i k [n] so that: σ(i) = i 1, σ 2 (i) = i 2,..., σ k (i) = i k = i Proof. Because [n] is finite, the eements of the orbit i, σ(i), σ 2 (i),... eventuay repeat. Suppose the first repetition is: σ m (i) = σ n (i) with m < n Then composing both sides with the permutation (σ m ) 1 = (σ 1 ) m gives i = σ n m (i), and the Proposition hods with k = n m. Cyce Notation. The cyce notation for σ Perm(n) ists the distinct eements of the orbit of 1 (in parentheses without commas), foowed by the distinct eements of the orbit of the first eement not contained in the orbit of 1, etc. unti a eements of [n] are exhausted. For exampe, in cyce notation the eements of Perm(3) are: (1, 2, 3) = (1)(2)(3) (1, 3, 2) = (1)(2 3), (2, 1, 3) = (1 2)(3), (3, 2, 1) = (1 3)(2) (2, 3, 1) = (1 2 3), (3, 1, 2) = (1 3 2) Simpification. Singeton orbits are eft out of the cyce notation, with the assumption that any missing eement is fixed by the permutation. For exampe: (1, 3, 2) = (1)(2 3) = (2 3) in the simpified notation.

5 5 The composition tabe for the six eements of Perm(3) is: g f g = id (1 2) (1 3) (2 3) (1 2 3) (1 3 2) f = id id (1 2) (1 3) (2 3) (1 2 3) (1 3 2) (1 2) (1 2) id (1 2 3) (1 3 2) (1 3) (2 3) (1 3) (1 3) (1 3 2) id (1 2 3) (2 3) (1 2) (2 3) (2 3) (1 2 3) (1 3 2) id (1 2) (1 3) (1 2 3) (1 2 3) (2 3) (1 2) (1 3) (1 3 2) id (1 3 2) (1 3 2) (1 3) (2 3) (1 2) id (1 2 3) Exampe 1.3. The 24 eements of Perm(4) fit into one of five casses: (i) The identity id (ii) Transpositions (iii) Three-cyces (1 2), (1 3), (1 4), (2 3), (2 4), (3 4) (1 2 3), (1 3 2), (1 2 4), (1 4 2), (1 3 4), (1 4 3), (2 3 4), (2 4 3) (iv) Four-cyces ( ), ( ), ( ), ( ), ( ), ( ) (v) Transposition pairs (1 2)(3 4), (1 3)(2 4), (1 4)(2 3) We eave the composition tabe for Perm(4) as an exercise. Exercise 1.1. The sign of a transposition (for any n) is: sgn(i j) = 1 Coroary 1.1. The sign of an m-cyce is ( 1) m 1. Proof. An m-cyce is a composition of m 1 transpositions: (i 1 i 2 i 3... i m ) = (i 1 i m ) (i 1 i 3 ) (i 1 i 2 ) so by the Exercise and Prop 1.1, we have sgn(i 1 i 2 i 3... i m ) = ( 1) m 1. Coroary 1.2. Every permutation σ Perm(n) is a composition of transpositions. Athough the number of such transpositions is not we-defined, the parity (even or odd) of the number is we-defined. Proof. The cyce notation presents a permutation as a composition of (disjoint) cyces. Each cyce is a composition of transpositions, as in the Proof of Coroary 1.1 above. The parity is determined by the sign of σ, which was we-defined in Definition 1.2.

6 6 Exercises Prove that the sign of a transposition (i j) Perm(n) is Work out the composition tabe for Perm(4). A subset of Perm(n) that contains the identity id [n] and is cosed under inverses and compositions is a subgroup Find a the subgroups of Perm(3) and Perm(4). Hint: The number of eements in a subgroup of Perm(n) divides n!. The order of σ Perm(n) is the smaest vaue of k so that σ k = id [n] (a) Obtain the order of σ Perm(n) from its cyce notation. (Concude that the order of any permutation is finite!) (b) Find the argest order of an eement of Perm(n) for sma n. A nonempty set S is finite if there is a surjective map f : [n] S (a) If S is finite and non-empty, contempate why there is a bijective map f : [m] S for a unique integer m. (b) Find two infinite sets that are not isomorphic Let S be a fixed set and consider the foowing pair of coections: (a) The coection of subsets of S, (b) Incusions of subsets This is a category! Draw pictures of it when S = [2] and [3]. What are compositions and isomorphisms in this category?