B. F. Caviness

Quantifier elimination and cylindrical algebraic decomposition

1 Introduction to the Method.- 2 Importance of QE and CAD Algorithms.- 3 Alternative Approaches.- 4 Practical Issues.- Acknowledgments.- Quantifier Elimination by Cylindrical Algebraic Decomposition - Twenty Years of Progress.- 1 Introduction.- 2 Original Method.- 3 Adjacency and Clustering.- 4 Improved Projection.- 5 Partial CADs.- 6 Interactive Implementation.- 7 Solution Formula Construction.- 8 Equational Constraints.- 9 Subalgorithms.- 10 Future Improvements.- A Decision Method for Elementary Algebra and Geometry.- 1 Introduction.- 2 The System of Elementary Algebra.- 3 Decision Method for Elementary Algebra.- 4 Extensions to Related Systems.- 5 Notes.- 6 Supplementary Notes.- Quantifier Elimination for Real Closed Fields by Cylindrical Algebraic Decomposition.- 1 Introduction.- 2 Algebraic Foundations.- 3 The Main Algorithm.- 4 Algorithm Analysis.- 5 Observations.- Super-Exponential Complexity of Presburger Arithmetic.- 1 Introduction and Main Theorems.- 2 Algorithms.- 3 Method for Complexity Proofs.- 4 Proof of Theorem 3 (Real Addition).- 5 Proof of Theorem 4 (Lengths of Proofs for Real Addition).- 6 Proof of Theorems 1 and 2 (Presburger Arithmetic).- 7 Other Results.- Cylindrical Algebraic Decomposition I: The Basic Algorithm.- 1 Introduction.- 2 Definition of Cylindrical Algebraic Decomposition.- 3 The Cylindrical Algebraic Decomposition Algorithm: Projection Phase.- 4 The Cylindrical Algebraic Decomposition Algorithm: Base Phase.- 5 The Cylindrical Algebraic Decomposition Algorithm: Extension Phase.- 6 An Example.- Cylindrical Algebraic Decomposition II: An Adjacency Algorithm for the Plane.- 1 Introduction.- 2 Adjacencies in Proper Cylindrical Algebraic Decompositions.- 3 Determination of Section-Section Adjacencies.- 4 Construction of Proper Cylindrical Algebraic Decompositions.- 5 An Example.- An Improvement of the Projection Operator in Cylindrical Algebraic Decomposition.- 1 Introduction.- 2 Idea.- 3 Analysis.- 4 Empirical Results.- Partial Cylindrical Algebraic Decomposition for Quantifier Elimination.- 1 Introduction.- 2 Main Idea.- 3 Partial CAD Construction Algorithm.- 4 Strategy for Cell Choice.- 5 Illustration..- 6 Empirical Results.- 7 Conclusion.- Simple Solution Formula Construction in Cylindrical Algebraic Decomposition Based Quantifier Elimination.- 1 Introduction.- 2 Problem Statement.- 3 (Complex) Solution Formula Construction.- 4 Simplification of Solution Formulas.- 5 Experiments.- Recent Progress on the Complexity of the Decision Problem for the Reals.- 1 Some Terminology.- 2 Some Complexity Highlights.- 3 Discussion of Ideas Behind the Algorithms.- An Improved Projection Operation for Cylindrical Algebraic Decomposition.- 1 Introduction..- 2 Background Material..- 3 Statements of Theorems about Improved Projection Map.- 4 Proof of Theorem 3 (and Lemmas).- 5 Proof of Theorem 4 (and Lemmas).- 6 CAD Construction Using Improved Projection.- 7 Examples.- 8 Appendix.- Algorithms for Polynomial Real Root Isolation.- 1 Introduction.- 2 Preliminary Mathematics.- 3 Algorithms.- 4 Computing Time Analysis.- 5 Empirical Computing Times.- Sturm- Twenty Years of Progress.- 1 Introduction.- 2 Original Method.- 3 Adjacency and Clustering.- 4 Improved Projection.- 5 Partial CADs.- 6 Interactive Implementation.- 7 Solution Formula Construction.- 8 Equational Constraints.- 9 Subalgorithms.- 10 Future Improvements.- A Decision Method for Elementary Algebra and Geometry.- 1 Introduction.- 2 The System of Elementary Algebra.- 3 Decision Method for Elementary Algebra.- 4 Extensions to Related Systems.- 5 Notes.- 6 Supplementary Notes.- Quantifier Elimination for Real Closed Fields by Cylindrical Algebraic Decomposition.- 1 Introduction.- 2 Algebraic Foundations.- 3 The Main Algorithm.- 4 Algorithm Analysis.- 5 Observations.- Super-Exponential Complexity of Presburger Arithmetic.- 1 Introduction and Main Theorems.- 2 Algorithms.- 3 Method for Complexity Proofs.- 4 Proof of Theorem 3 (Real Addition).- 5 Proof of Theorem 4 (Lengths of Proofs for Real Addition).- 6 Proof of Theorems 1 and 2 (Presburger Arithmetic).- 7 Other Results.- Cylindrical Algebraic Decomposition I: The Basic Algorithm.- 1 Introduction.- 2 Definition of Cylindrical Algebraic Decomposition.- 3 The Cylindrical Algebraic Decomposition Algorithm: Projection Phase.- 4 The Cylindrical Algebraic Decomposition Algorithm: Base Phase.- 5 The Cylindrical Algebraic Decomposition Algorithm: Extension Phase.- 6 An Example.- Cylindrical Algebraic Decomposition II: An Adjacency Algorithm for the Plane.- 1 Introduction.- 2 Adjacencies in Proper Cylindrical Algebraic Decompositions.- 3 Determination of Section-Section Adjacencies.- 4 Construction of Proper Cylindrical Algebraic Decompositions.- 5 An Example.- An Improvement of the Projection Operator in Cylindrical Algebraic Decomposition.- 1 Introduction.- 2 Idea.- 3 Analysis.- 4 Empirical Results.- Partial Cylindrical Algebraic Decomposition for Quantifier Elimination.- 1 Introduction.- 2 Main Idea.- 3 Partial CAD Construction Algorithm.- 4 Strategy for Cell Choice.- 5 Illustration..- 6 Empirical Results.- 7 Conclusion.- Simple Solution Formula Construction in Cylindrical Algebraic Decomposition Based Quantifier Elimination.- 1 Introduction.- 2 Problem Statement.- 3 (Complex) Solution Formula Construction.- 4 Simplification of Solution Formulas.- 5 Experiments.- Recent Progress on the Complexity of the Decision Problem for the Reals.- 1 Some Terminology.- 2 Some Complexity Highlights.- 3 Discussion of Ideas Behind the Algorithms.- An Improved Projection Operation for Cylindrical Algebraic Decomposition.- 1 Introduction..- 2 Background Material..- 3 Statements of Theorems about Improved Projection Map.- 4 Proof of Theorem 3 (and Lemmas).- 5 Proof of Theorem 4 (and Lemmas).- 6 CAD Construction Using Improved Projection.- 7 Examples.- 8 Appendix.- Algorithms for Polynomial Real Root Isolation.- 1 Introduction.- 2 Preliminary Mathematics.- 3 Algorithms.- 4 Computing Time Analysis.- 5 Empirical Computing Times.- Sturm-Habicht Sequences, Determinants and Real Roots of Univariate Polynomials.- 1 Introduction.- 2 Algebraic Properties of Sturm-Habicht Sequences.- 3 Sturm-Habicht Sequences and Real Roots of Polynomial.- 4 Sturm-Habicht Sequences and Hankel Forms.- 5 Applications and Examples.- Characterizations of the Macaulay Matrix and Their Algorithmic Impact.- 1 Introduction.- 2 Notation.- 3 Definitions of the Macaulay Matrix.- 4 Extraneous Factor and First Properties of the Macaulay Determinant.- 5 Characterization of the Macaulay Matrix.- 6 Characterization of the Macaulay Matrix, if It Is Used to Calculate the u-Resultant.- 7 Two Sorts of Homogenization.- 8 Characterization of the Matrix of the Extraneous Factor.- 9 Conclusion.- Computation of Variant Resultants.- 1 Introduction.- 2 Problem Statement.- 3 Review of Determinant Based Method.- 4 Quotient Based Method.- 5 Modular Methods..- 6 Theoretical Computing Time Analysis.- 7 Experiments.- A New Algorithm to Find a Point in Every Cell Defined by a Family of Polynomials.- 1 Introduction.- 2 Proof of the Theorem.- Local Theories and Cylindrical Decomposition.- 1 Introduction.- 2 Infinitesimal Sectors at the Origin.- 3 Neighborhoods of Infinity.- 4 Exponential Polynomials in Two Variables.- A Combinatorial Algorithm Solving Some Quantifier Elimination Problems.- 1 Introduction.- 2 Sturm-Habicht Sequence.- 3 The Algorithms.- 4 Conclusions.- A New Approach to Quantifier Elimination for Real Algebra.- 1 Introduction.- 2 The Quantifier Elimination Problem for the Elementary Theory of the Reals.- 3 Counting Real Zeros Using Quadratic Forms.- 4 Comprehensive Grobner Bases.- 5 Steps of the Quantifier Elimination Method.- 6 Examples.- References.

On Canonical Forms and Simplification

This paper deals with the simplification problem of symbolic mathematics. The notion of canonical form is defined and presented as a well-defined alternative to the concept of simplified form. Following Richardson it is shown that canonical forms do not exist for sufficiently rich classes of mathematical expressions. However, with the aid of a nmnber- theoretic conjecture, a large subclass of the negative classes is shown to possess a canonical form.

An Extension of Liouville’s Theorem on Integration in Finite Terms

In Part I of this paper, we give an extension of Liouville’s Theorem and give a number of examples which show that integration with special functions involves some phenomena that do not occur in integration with the elementary functions alone. Our main result generalizes Liouville’s Theorem by allowing, in addition to the elementary functions, special functions such as the error function, Fresnel integrals and the logarithmic integral (but not the dilogorithm or exponential integral) to appear in the integral of an elementary function. The basic conclusion is that these functions, if they appear, appear linearly. We give an algorithm which decides if an elementary function, built up using only exponential functions and rational operations has an integral which can be expressed in terms of elementary functions and error functions.

A Structure Theorem for Exponential and Primitive Functions

In this paper a new theorem is proved that generalizes a result of Risch. The new theorem gives all the possible algebraic relationships among functions that can be built up from the rational functions by algebraic operations, by taking exponentials, and by integration. The functions so generated are called exponential and primitive functions. From the theorem an algorithm for determining algebraic dependence among a given set of exponential and primitive functions is derived. The algorithm is then applied to a problem in computer algebra.

Integration in Finite Terms with Special Functions: A Progress Report

Since R. Risch published an algorithm for calculating symbolic integrals of elementary functions in 1969, there has been an interest in extending his methods to include nonelementary functions. We report here on the recent development of two decision procedures for calculating integrals of transcendental elementary functions in terms of logarithmic integrals and error functions. Both of these algorithms are based on the Singer, Saunders, Caviness extension of Liouvilles theorem on integration in finite terms [Ssc81]. Parts of the logarithmic integral algorithm have been implemented in Macsyma and a brief demonstration is given.

A liouville theorem on integration in finite terms for line integrals

A multivariate generalization for line integrals of the Strong Liouville Theorem due to Risch is presented. The result is an abstract version of the following: Let k be a subfield of the field of complex numbers. Let each , be any function in a field E obtained by algebraic operations and the taking of logarithms and exponentials over . If there exists a functions g obtained by algebraic operations and the taking of logarithms and exponentials of elements of E such that then g must be of the form where do is in E, the ci are constants in k(a), and the di are elements in E(a), where a is a constant algebraic over E.

A structure theorem for the elementary functions and its application to the identity problem

This paper uses elementary algebraic methods to obtain new proofs for theorems on algebraic relationships between the logarithmic and exponential functions. The main result is a multivariate version of a special case of the structure theorem due to Risch that gives, in a very explicit fashion, the possible algebraic relationships between the exponential and logarithm functions. The structure theorem has important applications to symbolic mathematical computation in that it in essence provides a canonical form for the elementary transcendental functions, and hence solves the identity problem for this class of functions. Such applications are discussed in the last section.

An extension of Liouville's theorem on integration in finite terms

In this paper we give an extension of the Liouville theorem [RISC69, p. 169] and give a number of examples which show that integration with special functions involves some phenomena that do not occur in integration with the elementary functions alone. Our main result generalizes Liouvilles theorem by allowing, in addition to the elementary functions, special functions such as the error function, Fresnel integrals and the logarithmic integral to appear in the integral of an elementary function. The basic conclusion is that these functions, if they appear, appear linearly.