V. Fields and Galois Theory

Similar documents
Existentially closed models of the theory of differential fields with a cyclic automorphism

Projective Lattices. with applications to isotope maps and databases. Ralph Freese CLA La Rochelle

CARDINALITIES OF RESIDUE FIELDS OF NOETHERIAN INTEGRAL DOMAINS

COMBINATORICS OF REDUCTIONS BETWEEN EQUIVALENCE RELATIONS

Gödel algebras free over finite distributive lattices

Double Ore Extensions versus Iterated Ore Extensions

Algebra and Number Theory Exercise Set

maps 1 to 5. Similarly, we compute (1 2)(4 7 8)(2 1)( ) = (1 5 8)(2 4 7).

INFLATION OF FINITE LATTICES ALONG ALL-OR-NOTHING SETS TRISTAN HOLMES J. B. NATION

3 The Model Existence Theorem

Orthogonality to the value group is the same as generic stability in C-minimal expansions of ACVF

REMARKS ON K3 SURFACES WITH NON-SYMPLECTIC AUTOMORPHISMS OF ORDER 7

Lecture l(x) 1. (1) x X

Laurence Boxer and Ismet KARACA

CS792 Notes Henkin Models, Soundness and Completeness

Transcendental lattices of complex algebraic surfaces

Introduction Ideal lattices Ring-SIS Ring-LWE Other algebraic lattices Conclusion. Ideal Lattices. Damien Stehlé. ENS de Lyon. Berkeley, 07/07/2015

Laurence Boxer and Ismet KARACA

Introduction to Priestley duality 1 / 24

THE IRREDUCIBILITY OF CERTAIN PURE-CYCLE HURWITZ SPACES

Worksheet A ALGEBRA PMT

A Property Equivalent to n-permutability for Infinite Groups

Optimal Stopping. Nick Hay (presentation follows Thomas Ferguson s Optimal Stopping and Applications) November 6, 2008

Wada s Representations of the. Pure Braid Group of High Degree

Integrating rational functions (Sect. 8.4)

Some Remarks on Finitely Quasi-injective Modules

A relation on 132-avoiding permutation patterns

The Factor Sets of Gr-Categories of the Type (Π,A)

Non replication of options

A DNC function that computes no effectively bi-immune set

Lie Algebras and Representation Theory Homework 7

The rth moment of a real-valued random variable X with density f(x) is. x r f(x) dx

The Turing Definability of the Relation of Computably Enumerable In. S. Barry Cooper

Optimal Allocation of Policy Limits and Deductibles

Palindromic Permutations and Generalized Smarandache Palindromic Permutations

RUDIN-KEISLER POSETS OF COMPLETE BOOLEAN ALGEBRAS

Math 546 Homework Problems. Due Wednesday, January 25. This homework has two types of problems.

Fuzzy L-Quotient Ideals

THE NUMBER OF UNARY CLONES CONTAINING THE PERMUTATIONS ON AN INFINITE SET

The finite lattice representation problem and intervals in subgroup lattices of finite groups

EXTENSIVE AND NORMAL FORM GAMES

A discretionary stopping problem with applications to the optimal timing of investment decisions.

arxiv: v1 [q-fin.pm] 13 Mar 2014

Logarithmic derivatives of densities for jump processes

Modular and Distributive Lattices

MATH 121 GAME THEORY REVIEW

The Real Numbers. Here we show one way to explicitly construct the real numbers R. First we need a definition.

Algebra homework 8 Homomorphisms, isomorphisms

INTERVAL DISMANTLABLE LATTICES

Game Theory: Normal Form Games

Martingales. by D. Cox December 2, 2009

Harvard School of Engineering and Applied Sciences CS 152: Programming Languages

Part 3: Trust-region methods for unconstrained optimization. Nick Gould (RAL)

On the action of the absolute Galois group on triangle curves

On equation. Boris Bartolomé. January 25 th, Göttingen Universität & Institut de Mathémathiques de Bordeaux

Introduction to Probability Theory and Stochastic Processes for Finance Lecture Notes

Edexcel past paper questions. Core Mathematics 4. Binomial Expansions

A1: American Options in the Binomial Model

Computing Unsatisfiable k-sat Instances with Few Occurrences per Variable

CATEGORICAL SKEW LATTICES

Hyperidentities in (xx)y xy Graph Algebras of Type (2,0)

Approximating the Transitive Closure of a Boolean Affine Relation

Notes on the symmetric group

Microeconomic Theory II Preliminary Examination Solutions

Expectations. Definition Let X be a discrete rv with set of possible values D and pmf p(x). The expected value or mean value of X, denoted by E(X ) or

Theorem 1.3. Every finite lattice has a congruence-preserving embedding to a finite atomistic lattice.

Chapter 4 Partial Fractions

SYLLABUS AND SAMPLE QUESTIONS FOR MSQE (Program Code: MQEK and MQED) Syllabus for PEA (Mathematics), 2013

Research Article On the Classification of Lattices Over Q( 3) Which Are Even Unimodular Z-Lattices of Rank 32

On Complexity of Multistage Stochastic Programs

Math-Stat-491-Fall2014-Notes-V

Simplicity of associative and non-associative Ore extensions

Introduction Random Walk One-Period Option Pricing Binomial Option Pricing Nice Math. Binomial Models. Christopher Ting.

Inversion Formulae on Permutations Avoiding 321

Quadrant marked mesh patterns in 123-avoiding permutations

Lecture 2: The Simple Story of 2-SAT

Harvard School of Engineering and Applied Sciences CS 152: Programming Languages

3.1 Factors and Multiples of Whole Numbers

On the generalized σ-fitting subgroup of finite groups

Collinear Triple Hypergraphs and the Finite Plane Kakeya Problem

R. Selvi 1, P. Thangavelu 2. Sri Parasakthi College for Women Courtallam, INDIA 2 Department of Mathematics

Continuous images of closed sets in generalized Baire spaces ESI Workshop: Forcing and Large Cardinals

MATH 220: DUHAMEL S PRINCIPLE

Qualifying Exam Solutions: Theoretical Statistics

Strong normalisation and the typed lambda calculus

MATH 116: Material Covered in Class and Quiz/Exam Information

A Translation of Intersection and Union Types

Exact Sampling of Jump-Diffusion Processes

ADDING A LOT OF COHEN REALS BY ADDING A FEW II. 1. Introduction

Version A. Problem 1. Let X be the continuous random variable defined by the following pdf: 1 x/2 when 0 x 2, f(x) = 0 otherwise.

GUESSING MODELS IMPLY THE SINGULAR CARDINAL HYPOTHESIS arxiv: v1 [math.lo] 25 Mar 2019

The Clarke Tax Algorithm

An Optimal Algorithm for Calculating the Profit in the Coins in a Row Game

Arbitrage Bounds for Volatility Derivatives as Free Boundary Problem. Bruno Dupire Bloomberg L.P. NY

Harvard School of Engineering and Applied Sciences CS 152: Programming Languages

START HERE: Instructions. 1 Exponential Family [Zhou, Manzil]

KAPLANSKY'S PROBLEM ON VALUATION RINGS

Optimal Stopping Rules of Discrete-Time Callable Financial Commodities with Two Stopping Boundaries

Permutation Factorizations and Prime Parking Functions

IEOR E4703: Monte-Carlo Simulation

Transcription:

Math 201C - Alebra Erin Pearse V.2. The Fundamental Theorem. V. Fields and Galois Theory 4. What is the Galois roup of F = Q( 2, 3, 5) over Q? Since F is enerated over Q by {1, 2, 3, 5}, we need to determine all possible imaes σ( 2), σ( 3), σ( 5) 1 for σ AutQF. Now 2 has irreducible polynomial f(x) = x 2 2 Q[x] 3 has irreducible polynomial (x) = x 2 3 Q[x] 5 has irreducible polynomial h(x) = x 2 5 Q[x] Since σ must take roots of f to roots of f (etc.), we must have ( ) σ 2 = ± ( ) 2 σ 3 = ± ( ) 3 σ 5 = ± 5 Since we have three independent binary choices, AutQF = Z 2 Z 2 Z 2. To verify this formally, note that AutQF = 8 and every element u of AutQF has order u 2. 6. Let f K(x) with f of K by K(x). / K and f, relatively prime in K[x] and consider the extension a) x is alebraic over K( f ) and [K(x) : K( f )] = max (de f, de ). Since f/ is transcendental over K, we write f/ = z and consider z as an indeterminate. Let the polynomial ϕ K(z)[y] be defined by ϕ (y) = z (y) f (y) so that x is clearly a root of ϕ. First we show that it is primitive and irreducible over K[z][y] = K[z, y], and then use III.6.13. ϕ is irreducible in K[z][y]. Suppose ϕ were reducible: but then < ϕ (y) = (z 1 (y) f 1 (y)) h (where f 1 h = f and 1 h = ), f, are relatively prime. ϕ is primitive. Suppose not: then for ϕ (y) = (zb 0 a 0 ) + (zb 1 a 1 ) y +... + (zb n a n ) y n there would be r K such that zb i a i = b i (z r), which implies a i = rb i, i. But this would mean f = r K, < hypothesis. Now that ϕ is a primitive irreducible polynomial, we know ϕ is irreducible over K(z)[y] by III.6.13. Hence, ϕ is the irreducible polynomial of x and x is thus alebraic over K( f ). By examination of ϕ, it is also clear that [K(x) : K( f )] = de ϕ = max (de f, de ). 1 Since σ fixes K, we know that we must have σ(1) = 1.

2 June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse b) If E K is an intermediate field of the extension K(x) : K, then [K(x) : E] <. E K, so pick u E\K. Then u E implies that u is of the form f(x), for some (x) f, K[x]\K. x is alebraic over K(u) by part (a), because E K = f / K and we can choose f, coprime. Let h be the irreducible polynomial of x over K(u). Then we have a tower of fields K (x) E K (h) K and a correspondin product of dimensions [K (x) : K (u)] = [K (x) : E] [E : K (u)]. Since [K (x) : K (u)] = de h <, this clearly implies [K(x) : E] <. c) x f ϕ(x) induces a homomorphism such that σ : ϕ (f/)/ψ (f/). ψ(x) σ Aut K F max (de f, de ) = 1. To see that σ induces the iven homomorphism, use the proof of 2.2. Note that σ is a field homomorphism, so it is certainly injective by III.2.21(iv). Now, [ ( )] f max (de f, de ) = 1 K (x) : K ( ) = 1 by (a) K (x) = K = σ (K (x)) d) Aut K K(x) = {x ax+b cx+d First note that the map ϕ(x) f and thus fixes K. Then f σ is surjective.. a, b, c, d K and ad bc 0} = GL 2 (K). ψ(x) ϕ(f/) ψ(f/) σ Aut K K(x) σ GL 2 (K) follows immediately from (c) and the fact that For a K\{0}, b K, the maps σ a : K(x) K(x) by σ a : f(x) (x) σ is invertible ad bc 0. f(ax) (ax) and are clearly K-automorphisms of K(x). discussed previously is just evaluation at τ b : K(x) K(x) by τ b : f(x) (x) f(x+b) (x+b)

June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse 3 9. a) If K is infinite, then K(x) is Galois over K. If K(x) is not Galois over K, then K F (Aut K K(x)), so we must have K E = F (Aut K K(x)). Then [K(x) : E] < by 6(b), which implies that [ AutE K(x) : Aut K(x) K(x) ] = [Aut E K(x) : 1] = Aut E K(x) [K(x) : E] by Lemma 2.8 < By hypothesis, K =, so Aut K K(x) = { ax+b cx+d} = by 6(d). But then Lemma 2.6(iv) ives which then implies < above. Aut K K(x) = Aut E K(x) Aut E K(x) = b) If K is finite, then K(x) is not Galois over K. First note that, as above, Aut K K(x) = ax+b cx+d by 6(d), so K = n < = Aut K K(x) n 4 < By hypothesis, [Aut K K(x) : 1 K ] <, so [F (1 K ) : F (Aut K K(x))] [Aut K K(x) : 1 K ] by Lemma 2.9. If K(x) were Galois, this would imply < x is transcendental. [K(x) : K] Aut K K(x) n 4 <

4 June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse 11. In the extension of Q by Q(x), show that the intermediate field Q(x 2 ) is closed, but Q(x 3 ) is not. a) Q(x 2 ) is closed u Q(x)\Q(x 2 ), σ AutQ(x 2 )Q(x) such that σ(u) u. Note that Q(x) is alebraic over Q(x 2 ) with basis {1, x}, because x is a root of f(y) = y 2 x 2 Q(x 2 )[y]. Then pick u Q(x)\Q(x 2 ), so u = a + bx where b 0. Let σ be defined by σ(1) = 1, σ(x) = x. Now σ(u) = a bx u, but for any v Q(x 2 ), define v = n i=0 a ix 2i so that σ(v) = n σ (a i) σ (x) 2i i=0 Thus, σ fixes Q(x 2 ) but not u. = n a i ( x) 2i i=0 = n a ix 2i i=0 = v b) To see that Q(x 3 ) is not closed, we will exhibit u Q(x)\Q(x 3 ) which is fixed by any σ AutQ(x 3 )Q(x). Let u = x + x 3 so that σ(u) = σ(x) + x 3 because σ fixes x 3 Q(x 3 ). We note that Q(x) is alebraic over Q(x 3 ) with basis {1, x, x 2 }, because x is a root of f(y) = y 3 x 3 Q(x 3 )[y]. So σ is completely determined by its action on {1, x}. We know that σ(1) = 1 as above (because σ must fix Q(x 3 )), but what is σ(x)? σ(x) must be a root of f(y), but f(y) has only one root: x. Thus, σ(x) = x = σ(u) = x + x 3 = u, for any σ which fixes Q(x 3 ). Thus, u F ( AutQ(x 3 )Q(x) ) \Q(x 3 ) = F ( AutQ(x 3 )Q(x) ) Q(x 3 ), i.e., Q(x 3 ) is not closed.

June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse 5 14. Let F : K be a finite-dimensional Galois extension with intermediate fields L, M. a) Aut LM F = Aut L F Aut M F LM is the smallest field containin L and M means that L LM, M LM and L, M E = LM E We proceed to show the equality by a double-inclusion arument. L LM = Aut LM F Aut L F, and M LM = Aut LM F Aut M F, so clearly Aut LM F Aut L F Aut M F. Makin extensive use of the fact that F : K is f.d.., Aut L F Aut M F Aut L F, Aut M F = F (Aut L F ), F (Aut M F ) F (Aut L F Aut M F ) = L, M F (Aut L F Aut M F ) Lemma 2.10 = LM F (Aut L F Aut M F ) by first line = Aut F(AutL F Aut M F )F Aut LM F by Fun. Thm. = Aut L F Aut M F Aut LM F Lemma 2.6(iii) b) Aut L T M F = Aut L F Aut M F L M L, M = Aut L F, Aut M F Aut L M F = Aut L F Aut M F Aut L M F, because the join is the smallest roup containin both Aut L F and Aut M F. F (Aut L M F ) = L M by f.d.., so Aut L F, Aut M F Aut L F Aut M F def of = F (Aut L F Aut M F ) F (Aut L F, Aut M F ) by Fun. Thm. = F (Aut L F Aut M F ) L, M by f.d.. = F (Aut L F Aut M F ) L M c) What conclusion can be drawn if Aut L F Aut M F = 1? In this case, Aut L F Aut M F = Aut LM F = 1 by above = F (Aut LM F = F ) by Fun. Thm. = LM = F by f.d..

6 June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse 15. a) F : K is a finite-dimensional Galois extension with intermediate field E =!L where L is the smallest field such that E L F and L is Galois over K. Let {L i } be the set of Galois extensions of K which contain E. Since [F : K] <, there are a finite number of them. Now define L = n i=1 L i so that L is the smallest extension of K containin E. To see that L : K is Galois, consider the correspondin subroups Aut Li F. We have Aut Li F Aut K F for each L i, by the Fun. Thm., and n Aut L F = Aut Li F = Aut Li F i=1 follows by #14(b). Now we have n Aut Li F Aut K F i=1 by exercise I.5.3(a) (with a brief induction). Then Aut L F G = L : K is Galois by the Fun. Thm. aain.

June 7, 2002 Math 201C - Alebra - V.2 Erin Pearse 7 b) Furthermore, Aut L F = σ σ (Aut EF ) σ 1 where σ runs over all of Aut K F. To see Aut L F σ σ (Aut EF ) σ 1, Aut L F = σ (Aut L F ) σ 1 σ σ (Aut E F ) σ 1 σ σ (Aut E F ) σ 1 σ Aut K F Aut L F Aut K F Aut L F Aut E F Now to see σ σ (Aut EF ) σ 1 Aut L F. We know that Aut L F = Aut Li F, so it suffices to show that σ σ (Aut EF ) σ 1 Aut Li F for some i, i.e., that σ σ (Aut EF ) σ 1 is normal in Aut K F and contained in Aut E F. To see σ σ (Aut EF ) σ 1 Aut K F, pick τ Aut K F. Then ) τ ( σ σ (Aut EF ) σ 1 τ 1 = τ ( σ (Aut E F ) σ 1) τ 1 σ = σ τσ (Aut EF ) (τσ) 1 = ρ ρ (Aut EF ) ρ 1 = σ σ (Aut EF ) σ 1. To see why the last steps hold, consider that the intersection runs over all permutations σ Aut K F, but it doesn t matter in which order. Puttin στ = ρ amount to shufflin the indices and reorderin them in their oriinal state. Now σ σ (Aut EF ) σ 1 is invariant under conjuation and hence normal, which means that σ σ (Aut EF ) σ 1 = Aut Lj F for some j. Thus, σ σ (Aut EF ) σ 1 Aut Li F = Aut Li F = Aut L F.