<p>This thesis has three major topics. The first is on generalized multiplicities. The second is on height bounds for ideals of minors of matrices with a given rank. The last topic is on the ideal of minors of generic generalized diagonal matrices.</p> <p>In the first part of this thesis, we discuss various generalizations of Hilbert-Samuel multiplicity. These include the Buchsbaum-Rim multiplicity, mixed multiplicities, $j$-multiplicity, and $\varepsilon$-multiplicity. For $(R,m)$ a Noetherian local ring of dimension $d$ and $I$ a $m$-primary ideal in $R$, Lech showed the following bound for the Hilbert-Samuel multiplicity of $I$, $e(I) \leq d!\lambda(R/I)e(m)$. Huneke, Smirnov, and Validashti improved the bound to $e(mI) \leq d!\lambda(R/I)e(m)$. We generalize the improved bound to the Buchsbaum-Rim multiplicity and to mixed multiplicities. </p> <p>For the second part of the thesis we discuss bounds on heights of ideals of minors of matrices. A classical bound for these heights was shown by Eagon and Northcott. Bruns' bound is an improvement on the Eagon-Northcott bound taking into consideration the rank of the matrix. We prove an analogous bound to Bruns' bound for alternating matrices. We then discuss an open problem by Eisenbud, Huneke, and Ulrich that asks for height bounds for symmetric matrices given their rank. We show a few reduction steps and prove some small cases of this problem. </p> <p>Finally, for the last topic we explore properties of the ideal of minors of generic generalized diagonal matrices. Generalized diagonal matrices are matrices with two ladders of zeros in the bottom left and top right corners. We compute their initial ideals and give a description of the facets of their Stanley-Reisner complex. Using this description, we characterize when these ideals are Cohen-Macaulay. In the special case where the ladders of zeros are triangles, we compute the height and multiplicity</p>

We develop the necessary theory in algebraic geometry to place Bayesian networks into the realm of algebraic statistics. This allows us to create an algebraic geometry--statistics dictionary. In particular, we study the algebraic varieties defined by the conditional independence statements of Bayesian networks. A complete algebraic classification, in terms of primary decomposition of polynomial ideals, is given for Bayesian networks on at most five random variables. Hidden variables are related to the geometry of higher secant varieties. Moreover, a complete algebraic classification, in terms of generating sets of polynomial ideals, is given for Bayesian networks on at most three random variables and one hidden variable. The relevance of these results for model selection is discussed. / Ph. D.

The concept of unique factorization was first recognized in the 1840s, but even then, it was still fairly believed to be automatic. The error of this assumption was exposed largely through attempts to prove Pierre de Fermat's, 1601-1665, last theorem. Once mathematicians discovered that this property did not always hold, it was only natural for them to try to search for the strongest available alternative. Thus began the attempt to generalize unique factorization. Using the ascending chain condition on principle ideals, we will show the conditions under which a ring is a unique factorization domain.

This thesis deals with toric ideals associated with finite simple graphs. In particular we establish some results pertaining to the nature of the generators and syzygies of toric ideals associated with finite simple graphs. The first result dealt with in this thesis expands upon work by Favacchio, Hofscheier, Keiper, and Van Tuyl which states that for G, a graph obtained by "gluing" a graph H1 to a graph H2 along an induced subgraph, we can obtain the toric ideal associated to G from the toric ideals associated to H1 and H2 by taking their sum as ideals in the larger ring and saturating by a particular monomial f. Our contribution is to sharpen the result and show that instead of a saturation by f, we need only examine the colon ideal with f^2. The second result treated by this thesis pertains to graded Betti numbers of toric ideals of complete bipartite graphs. We show that by counting specific subgraphs one can explicitly compute a minimal set of generators for the corresponding toric ideals as well as minimal generating sets for the first two syzygy modules. Additionally we provide formulas for some of the graded Betti numbers. The final topic treated pertains to a relationship between the fundamental group the finite simple graph G and the associated toric ideal to G. It was shown by Villareal as well as Hibi and Ohsugi that the generators of a toric ideal associated to a finite simple graph correspond to the closed even walks of the graph G, thus linking algebraic properties to combinatorial ones. Therefore it is a natural question whether there is a relationship between the toric ideal associated to the graph G and the fundamental group of the graph G. We show, under the assumption that G is a bipartite graph with some additional assumptions, one can conceive of the set of binomials in the toric ideal with coprime terms, B(IG), as a group with an appropriately chosen operation ⋆ and establish a group isomorphism (B(IG), ⋆) ∼= π1(G)/H where H is a normal subgroup. We exploit this relationship further to obtain information about the generators of IG as well as bounds on the Betti numbers. We are also able to characterise all regular sequences and hence compute the depth of the toric ideal of G. We also use the framework to prove that IG = (⟨G⟩ : (e1 · · · em)^∞) where G is a set of binomials which correspond to a generating set of π1(G). / Thesis / Doctor of Philosophy (PhD)

An artinian standard graded algebra has the weak Lefschetz property if the multiplication by a general linear form induces maps of maximal rank between consecutive degree components. It has the strong Lefschetz property if the multiplication by powers of a general linear form also induce maps of maximal rank between the appropriate degree components. These properties are mainly studied for the constraints they place, when present, on the Hilbert series of the algebra. While the majority of research on the Lefschetz properties has focused on characteristic zero, we primarily consider the presence of the properties in positive characteristic. We study the Lefschetz properties by considering the prime divisors of determinants of critical maps. First, we consider monomial complete intersections in a finite number of variables. We provide two complements to a result of Stanley. We next consider monomial almost complete intersections in three variables. We connect the characteristics in which the weak Lefschetz property fails with the prime divisors of the signed enumeration of lozenge tilings of a punctured hexagon. Last, we study how perturbations of a family of monomial algebras can change or preserve the presence of the Lefschetz properties. In particular, we introduce a new strategy for perturbations rooted in techniques from algebraic geometry.

In this dissertation we will look at properties of two different posets from different perspectives. The first poset is the Rees product of the face lattice of the n-cube with the chain. Specifically we study the Möbius function of this poset. Our proof techniques include straightforward enumeration and a bijection between a set of labeled augmented skew diagrams and barred signed permutations which label the maximal chains of this poset. Because the Rees product of this poset is Cohen-Macaulay, we find a basis for the top homology group and a representation of the top homology group over the symmetric group both indexed by the set of labeled augmented skew diagrams. We also show that the Möbius function of the Rees product of a graded poset with the t-ary tree and the Rees product of its dual with the t-ary tree coincide. We discuss labelings for Rees and Segre products in general, particularly the Rees product of the face lattice of a polytope with the chain. We also look at cases where the Möbius function of a poset is equal to the permanent of a matrix and we consider local h-vectors for the barycentric subdivision of the n-cube. In each section we state open conjectures. The second poset in this dissertation is the Dowling lattice. In particular we look at the k = 1 case, that is, the partition lattice. We study inequalities on the flag vector of the partition lattice via a weighted boustrophedon transform and determine a more generalized version for the Dowling lattice. We generalize a determinantal formula of Niven and conclude with conjectures and avenues of study.

For D an integral domain with field of fractions K and E a subset of K, the ring Int (E,D) = {f e K[X]lf (E) C D} of integer-valued polynomials on E has been well studies. In particulare, when E is a finite subset of D, Chapman, Loper, and Smith, as well as Boynton and Klingler, obtained a bound on the number of elements needed to generate a finitely generated ideal of Ing (E, D) in terms of the corresponding bound for D. We obtain analogous results for Int (r) (E, D) - {f e K [X]lf(k) (E) c D for all 0 < k < r} , for finite E and fixed integer r > 1. These results rely on the work of Skolem [23] and Brizolis [7], who found ways to characterize ideals of Int (E, D) from the values of their polynomials at points in D. We obtain similar results for E = D in case D is local, Noetherian, one-dimensional, analytically irreducible, with finite residue field. / by Yuri Villanueva. / Thesis (Ph.D.)--Florida Atlantic University, 2012. / Includes bibliography. / Mode of access: World Wide Web. / System requirements: Adobe Reader.

This paper determines whether monoid rings with the two-generator property have the strong two-generator property. Dedekind domains have both the two-generator and strong two-generator properties. How common is this? Two cases are considered here: the zero-dimensional case and the one-dimensional case for monoid rings. Each case is looked at to determine if monoid rings that are not PIRs but are two-generated have the strong two-generator property. Full results are given in the zero-dimensional case, however only partial results have been found for the one-dimensional case.

The classical discriminant D_n of degree n polynomials detects whether a given univariate polynomial f has a repeated root. It is itself a polynomial in the coefficients of f which, according to several classical results by Bézout, Sylvester, Cayley, and others, may be expressed as the determinant of a matrix whose entries are much simpler polynomials in the coefficients of f. This thesis is concerned with the construction and classification of such determinantal formulae for D_n. In particular, all of the formulae for D_n appearing in the classical literature are equivalent in the sense that the cokernels of their associated matrices are isomorphic as modules over the associated polynomial ring. This begs the question of whether there exist formulae which are not equivalent to the classical formulae and not trivial in the sense of having the same cokernel as the 1 x 1 matrix (D_n). The results of this thesis lie in two directions. First, we construct an explicit non-classical formula: the presentation matrix of the open swallowtail first studied by Arnol'd and Givental. We study the properties of this formula, contrasting them with the properties of the classical formulae. Second, for the discriminant of the polynomial x^4+a_2x^2+a_3x+a_4 we embark on a classification of determinantal formulae which are homogeneous in the sense that the cokernels of their associated matrices are graded modules with respect to the grading deg a_i=i. To this end, we use deformation theory: a moduli space of such modules arises from the base spaces of versal deformations of certain modules over the E_6 singularity {x^4-y^3}. The method developed here can in principle be used to classify determinantal formulae for all discriminants, and, indeed, for all singularities.

The classical discriminant D_n of degree n polynomials detects whether a given univariate polynomial f has a repeated root. It is itself a polynomial in the coefficients of f which, according to several classical results by Bézout, Sylvester, Cayley, and others, may be expressed as the determinant of a matrix whose entries are much simpler polynomials in the coefficients of f. This thesis is concerned with the construction and classification of such determinantal formulae for D_n. In particular, all of the formulae for D_n appearing in the classical literature are equivalent in the sense that the cokernels of their associated matrices are isomorphic as modules over the associated polynomial ring. This begs the question of whether there exist formulae which are not equivalent to the classical formulae and not trivial in the sense of having the same cokernel as the 1 x 1 matrix (D_n). The results of this thesis lie in two directions. First, we construct an explicit non-classical formula: the presentation matrix of the open swallowtail first studied by Arnol'd and Givental. We study the properties of this formula, contrasting them with the properties of the classical formulae. Second, for the discriminant of the polynomial x^4+a_2x^2+a_3x+a_4 we embark on a classification of determinantal formulae which are homogeneous in the sense that the cokernels of their associated matrices are graded modules with respect to the grading deg a_i=i. To this end, we use deformation theory: a moduli space of such modules arises from the base spaces of versal deformations of certain modules over the E_6 singularity {x^4-y^3}. The method developed here can in principle be used to classify determinantal formulae for all discriminants, and, indeed, for all singularities.