Complex Polynomial Systems

Introduction

The Wolfram Language functions Reduce, Resolve, and FindInstance allow you to solve a wide variety of problems that can be expressed in terms of equations and inequalities. The functions use a collection of algorithms applicable to classes of problems satisfying particular properties, as well as a set of heuristics that attempt to reduce the given problem to a sequence of problems that can be solved using the algorithms. This tutorial describes the algorithms used to solve the class of problems known as complex polynomial systems. It characterizes the structure of the returned answers and describes the options that affect various aspects of the methods involved.

A complex polynomial system is an expression constructed with polynomial equations and inequations

combined using logical connectives and quantifiers

An occurrence of a variable inside or is called a bound occurrence, and any other occurrence of is called a free occurrence. A variable is called a free variable of a complex polynomial system if the system contains a free occurrence of . A complex polynomial system is quantifier-free if it contains no quantifiers.

Here is an example of a complex polynomial system with free variables , , and .

In the Wolfram Language, quantifiers are represented using the functions Exists () and ForAll ().

Any complex polynomial system can be transformed to the prenex normal form

where each is a quantifier or , and is quantifier-free.

Any quantifier-free complex polynomial system can be transformed to the disjunctive normal form

where each is a polynomial equation or inequation.

Reduce, Resolve, and FindInstance always put complex polynomial systems in the prenex normal form, with quantifier-free parts in the disjunctive normal form, and subtract the sides of the equations and inequations to put them in the form

In all the tutorials for complex polynomial system solving, assume that the system has been transformed to this form.

Reduce can solve arbitrary complex polynomial systems. The solution (possibly after expanding with respect to ) is a disjunction of terms of the form

where are the free variables of the system, each is a polynomial, each is an algebraic function expressed using radicals or Root objects, and any terms of the conjunction (2) may be absent. Each is well defined, that is, no denominators or leading terms of Root objects in become zero for any satisfying the preceding terms of the conjunction (2).

This solves the system (1):

Resolve can eliminate quantifiers from arbitrary complex polynomial systems. If no variables are specified, the result is a logical combination of terms

where and are polynomials, and each is a free variable of the system. With variables specified in the input, Resolve gives the same answer as Reduce.

This eliminates quantifiers from the system (1):

FindInstance can handle arbitrary complex polynomial systems giving instances of complex solutions, or an empty list for systems that have no solutions. If the number of instances requested is more than one, the instances are randomly generated from the full solution of the system, and therefore they may depend on the value of the RandomSeed option. If one instance is requested, a faster algorithm that produces one instance is used, and the instance returned is always the same.

This finds a solution for the system (1):

The main tool used in solving complex polynomial systems is the Gröbner basis algorithm [1], which is available in the Wolfram Language as the GroebnerBasis function.

Gröbner Bases

Theory

This section gives a very brief introduction to the theory of Gröbner bases. It presents only the properties that are necessary to describe the algorithms used by the Wolfram Language in solving complex polynomial systems. For a more complete presentation see, for example, [1, 2]. Note that what [2] calls a monomial, [1] calls a term, and vice versa. This tutorial uses the terminology of [1].

A monomial in is an expression of the form with nonnegative integers .

Let be the set of all monomials in . A monomial order is a linear order on , such that for all , and implies for all .

Let be a field, the domain of integers, or the domain of univariate polynomials over a field. Let and be functions defined as follows. If is a field, , and . If is the domain of integers, and are the integer quotient and remainder functions, with . If is the domain of univariate polynomials over a field, and are the polynomial quotient and remainder functions.

A product , where is a nonzero element of and is a monomial, is called a term.

Let be a monomial order on , and let . The leading monomial of is the -largest monomial appearing in , the leading coefficient of is the coefficient at in , and the leading term of is the product .

A Gröbner basis of an ideal in , with respect to a monomial order , is a finite set of polynomials, such that for each , there exists , such that divides . Every ideal has a Gröbner basis (see [1] for a proof).

Let , and let be a monomial. A term is reducible modulo , if divides , and . If is reducible modulo , the reduction of modulo is the polynomial

Note that if , then ; otherwise, .

Let , and let be an ordered finite subset of . is reducible modulo if contains a term reducible modulo an element of . The reduction of modulo is defined by the following procedure. While the set of terms of reducible modulo an element of is not empty, take the term with the -largest monomial, take the first , such that is reducible modulo , and replace the term in with . Note that the monomials of terms chosen in subsequent steps of the procedure form a -descending chain, and each monomial can appear at most times, where is the number of elements of , hence the procedure terminates.

A Gröbner basis is semi-reduced if for all , is not reducible modulo , and if is the domain of integers, .

The Wolfram Language function GroebnerBasis returns semi-reduced Gröbner bases. In the following discussion, all Gröbner bases are assumed to be semi-reduced. Note that this is not the same as reduced Gröbner bases defined in the literature, since here the basis polynomials are not required to be monic. For a fixed monomial order, every ideal has a unique reduced Gröbner basis. Semi-reduced Gröbner bases defined here are only unique up to multiplication by invertible elements of (see Property 2).

Property 1: Let be a Gröbner basis of an ideal in , and let . Then iff .

This is a simple consequence of the definitions.

Property 2: Let and be two Gröbner bases of an ideal with respect to the same monomial order , and suppose that elements of and are ordered by their leading monomials. Then , and for all , if is the domain of integers, ; otherwise, for some invertible element of .

Proof: If , then is reducible modulo or is reducible modulo . Hence the leading monomials of the elements of a Gröbner basis are all different. Without loss of generality, assume . For induction, fix and suppose that for all , for some invertible element of . If is the domain of integers, . Without loss of generality, assume . Since belongs to , there exists such that divides . Then , and so . If , then would be reducible modulo and also modulo , which is impossible, since is semi-reduced. Hence , and , and divides . Similarly, divides . Therefore, there exists an invertible element of , such that . If is the domain of integers, and are positive, and so . Let . Suppose . Since belongs to , must be divisible by , for some . Let and be the coefficients at in and . If is a field, the term of is reducible modulo , which contradicts the assumption that is semi-reduced. If is the domain of univariate polynomials over a field,

and so either is reducible modulo , or is reducible modulo , which contradicts the assumption that and are semi-reduced. Finally, let be the domain of integers. Since neither is reducible modulo nor is reducible modulo , and . Hence , which is impossible, since is divisible by . Therefore , and so . By induction on , for all , . If , then would be reducible modulo some , with , and hence would be reducible modulo . Therefore , which completes the proof of Property 2.

Property 3: Let be an ideal in , let , and let be a Gröbner basis of the ideal in . Then belongs to the radical of iff for an invertible element of .

If an ideal contains invertible elements of , GroebnerBasis always returns .

Proof: Note first that

belongs to the ideal for any nonnegative integer . Hence, if belongs to the radical of , then 1 belongs to . Since is a Gröbner basis of , it must contain an element whose leading coefficient divides 1. Hence is an invertible element of . Since is semi-reduced and divides any term, . Now suppose that for an invertible element of . Then 1 belongs to , and so

where each belongs to , and each belongs to . Hence comparing coefficients at powers of leads to the following equations modulo : , , for , and . Then, , for , and modulo . Therefore, belongs to the radical of , which completes the proof of Property 3.

The following more technical property is important for solving complex polynomial systems.

Property 4: Let be a Gröbner basis of an ideal in with a monomial order that makes monomials containing greater than monomials not containing , let be the element of with the lowest positive degree in , let be the leading coefficient of in , and let be all elements of that do not depend on . Then for any polynomial and any point if , , for , and , then .

Proof: Consider the pseudoremainder of the division of by as polynomials in .

Since and belong to , so does . By Property 1, reduction of by must yield zero. Since the degree of in is less than , cannot be reduced by any of the elements of that depend on . Hence

and so . Since , (1) implies that , which completes the proof of Property 4.

Wolfram Language Function GroebnerBasis

The Wolfram Language function GroebnerBasis finds semi-reduced Gröbner bases. This section describes GroebnerBasis options used in the solving of complex polynomial systems.

option name
default value
CoefficientDomainAutomaticthe type of objects assumed to be coefficients
MethodAutomaticthe method used to compute the basis
MonomialOrderLexicographicthe criterion used for ordering monomials

GroebnerBasis options used in the solving of complex polynomial systems.

CoefficientDomain

This option specifies the domain of coefficients. With the default Automatic setting, the coefficient domain is the field generated by numeric coefficients present in the input.

Integersthe domain of integers
InexactNumbers[prec]inexact numbers with precision prec
Polynomials[x]the domain of polynomials in
RationalFunctionsthe field of rational functions in variables not on the variable list given to GroebnerBasis
Rationalsthe field of rational numbers

Available settings for CoefficientDomain.

Note that the coefficient domain also depends on the setting of the Modulus option of GroebnerBasis. With Modulus->p, for a prime number , the coefficient domain is the field , or the field of rational functions over if CoefficientDomain->RationalFunctions.

Method

With the default setting Method->Automatic, GroebnerBasis normally uses a variant of the Buchberger algorithm. Another algorithm available is the Gröbner walk, which computes a Gröbner basis in an easier monomial order and then transforms it to the required harder monomial order. This is often faster than directly computing a Gröbner basis in the required order, especially if the input polynomials are known to be a Gröbner basis for the easier order. With the Method->Automatic setting, GroebnerBasis uses the Gröbner walk for the default CoefficientDomain->Rationals and MonomialOrder->Lexicographic.

GroebnerBasis[polys,vars,Method->{"GroebnerWalk","InitialMonomialOrder"->order1},MonomialOrder->order2]
find a Gröbner basis in order1 and use the Gröbner walk algorithm to transform it to a Gröbner basis in order2

Transforming Gröbner bases using the Gröbner walk algorithm.

MonomialOrder

This option specifies the monomial order. The value can be either one of the named monomial orders or a weight matrix. The following table gives conditions for .

Lexicographic
DegreeLexicographic
DegreeReverseLexicographic

Monomial orders.

Quantifier elimination needs an order in which monomials containing quantifier variables are greater than monomials not containing quantifier variables. The Lexicographic order satisfies this condition, but the following EliminationOrder usually leads to faster computations.

where denotes total degree, denotes free variables, denotes quantifier variables, and are monomials, and denotes the DegreeReverseLexicographic order.

Using EliminationOrder requires the GroebnerBasis syntax with elimination variables specified.

GroebnerBasis[polys,xvars,yvars,MonomialOrder->EliminationOrder]
find a Gröbner basis in EliminationOrder

Gröbner basis in elimination order.

By default, GroebnerBasis with MonomialOrder->EliminationOrder drops the polynomials that contain from the result, returning only basis polynomials in . To get all basis polynomials, the value of the system option EliminateFromGroebnerBasis from the GroebnerBasisOptions group must be changed. (The Wolfram Language changes the option locally in the quantifier elimination algorithm.) The option value can be changed with

SetSystemOptions["GroebnerBasisOptions"->{"EliminateFromGroebnerBasis"->False}].
option name
default value
"EliminateFromGroebnerBasis"Truewhether GroebnerBasis with MonomialOrder->EliminationOrder should remove polynomials containing elimination variables

System option EliminateFromGroebnerBasis.

This eliminates from . The answer is a polynomial whose zeros are the Zariski closure of the projection of the solution set of the two original equations on the plane:
The exact description of the projection of the solution set on the plane depends on all basis polynomials. Note that the second basis polynomial cannot be zero if or is zero:
This resets the system option to its default value:
Resolve gives the exact description of the projection of the solution set on the plane:

Decision Problems

A decision problem is a system with all variables existentially quantified, that is, a system of the form

where are all variables in . Solving a decision problem means deciding whether it is equivalent to True or to False, that is, deciding whether the quantifier-free system of polynomial equations and inequations has solutions.

Solving this decision problem proves that a quadratic equation with a zero determinant cannot have two different roots:

Given the identities

solving any decision problem can be reduced to solving a finite number of decision problems of the form

By Hilbert's Nullstellensatz and Property 3 of Gröbner bases

has complex solutions iff

with an arbitrary monomial order, is different than {1}.

This shows that has complex solutions:
This shows that has no complex solutions:

When the Wolfram Language solves a decision problem, the monomial order used by the GroebnerBasis computation is MonomialOrder->EliminationOrder, with {z} specified as the elimination variable list. This setting corresponds to the monomial ordering in which monomials containing z are greater than those that do not contain z, and the ordering of monomials not containing z is degree reverse lexicographic. If there is no inequation condition, there is no need to introduce z, and the Wolfram Language uses MonomialOrder->DegreeReverseLexicographic.

Quantifier Elimination

For any complex polynomial system there exists an equivalent quantifier-free complex polynomial system. This follows from Chevalley's theorem, which states that a projection of a quasi-algebraically constructible set (a solution set of a quantifier-free system of polynomial equations and inequations) is a quasi-algebraically constructible set [3]. Quantifier elimination is the procedure of finding a quantifier-free complex polynomial system equivalent to a given complex polynomial system. In the Wolfram Language, quantifier elimination for complex polynomial systems is done by Resolve. It is also used by Reduce and FindInstance as the first step in solving or finding instances of solutions of complex polynomial systems.

Eliminating quantifiers from this system gives a condition for quadratic equations to have at least two different zeros:

For complex polynomial systems the Wolfram Language uses the following quantifier elimination method. Given the identities

eliminating quantifiers from any complex polynomial system can be reduced to a finite number of single existential quantifier eliminations from systems of the form

To eliminate the quantifier from (1), the Wolfram Language first computes the Gröbner basis of equations

with a monomial order that makes monomials containing greater than monomials not containing .

The monomial order used is EliminationOrder, with specified as the elimination variable list and all basis polynomials kept.

If contains no polynomials that depend on , then a quantifier-free system equivalent to (1) can be obtained by equating all elements of to zero, and asserting that at least one coefficient of as a polynomial in is not equal to zero. Otherwise let be the element of with the lowest positive degree in , let be the leading coefficient of in , and let be all elements of that do not depend on . Now (1) can be split into a disjunction of two systems

and

To eliminate the quantifier from (2), the quantifier elimination procedure is called recursively. Since the ideal generated by strictly contains the ideal generated by , the Noetherian property of polynomial rings guarantees finiteness of the recursion.

If belongs to the radical of the ideal generated by , which is exactly when 1 belongs to

(3) is equivalent to False. Otherwise let

be the pseudoremainder of the division of by as polynomials in . Then (3) is equivalent to the quantifier-free system

To show that (3) implies (4), suppose that satisfies (3). Then and there exists , such that

Since and belong to the ideal generated by ,

and . Hence

which implies that

To show that (4) implies (3), suppose that satisfies (4). Then

Since is a polynomial of degree , and is a nonzero polynomial of degree less than , there is a root of such that divides but not for some . If were zero, then would divide , which is impossible because it would imply that divides . Therefore . Property 4 shows that for any polynomial . Since is a Gröbner basis of the ideal generated by ,

which completes the proof of correctness of the quantifier elimination algorithm.

This eliminates the quantifier from . Here , , and . Since is a nonzero constant, (2) is False and the equivalent quantifier-free system is given by (4). Since is a nonzero constant, (4) becomes :
This resets the system option to its default value:

Arbitrary Complex Polynomial Systems

FindInstance

FindInstance can handle arbitrary complex polynomial systems giving instances of complex solutions, or an empty list for systems that have no solutions. If the number of instances requested is more than one, the instances are randomly generated from the full solution of the system given by Reduce. If one instance is requested, a faster algorithm that produces one instance is used. Here is a description of the algorithm used to find a single instance, or prove that a system has no solutions.

If the system contains general quantifiers (), the quantifier elimination algorithm is used to eliminate the innermost quantifiers until the system contains only existential quantifiers () or is quantifier-free. Note that

has solutions if and only if has solutions, and if is a solution of , then is a solution of (1). Hence to find instances of solutions of systems containing only existential quantifiers it is enough to be able to find instances of quantifier-free systems. Moreover, is a solution of

if and only if it is a solution of one of the , with , so it is enough to show how to find instances of solutions of

First compute the GroebnerBasis of with MonomialOrder->EliminationOrder, eliminating the polynomials that depend on (if there is no inequation condition, is the GroebnerBasis of with MonomialOrder->DegreeReverseLexicographic). If contains 1, there are no solutions. Otherwise, compute a subset of of the highest cardinality among subsets strongly independent modulo the ideal generated by with respect to the degree reverse lexicographic order ([1], Section 9.3). Reorder so that , and compute the lexicographic order GroebnerBasis of the ideal generated by . To compute , the Wolfram Language uses the Gröbner walk algorithm.

For each of the variables , , select the polynomial with the smallest leading monomial among elements of that depend on and not on . Let be the leading coefficient of as a polynomial in . If depends on a variable that is not in , replace with the lexicographic order Gröbner basis of the ideal generated by and . The following shows that this operation keeps strongly independent modulo the ideal generated by . Hence, possibly after a finite (by the Noetherian property of polynomial rings) number of extensions of , the leading coefficient of depends only on , for all . For the set of polynomials , let be the set of common zeros of elements of . Both and have dimension , and , hence any -dimensional irreducible component of is also a component of . Since does not vanish on any irreducible component of , it does not vanish on any -dimensional irreducible component of . Therefore, the Gröbner basis of and contains a polynomial depending only on . Let . To find a solution of (2), pick its last coordinates so that . For all , , and so by Property 4 if , for , is chosen to be the first root of , then . Moreover, , because otherwise would belong to , which would imply that , which is impossible since divides .

To prove the correctness of the aforementioned algorithm, it must be shown that extending by that depend on a variable not in preserves strong independence of modulo the ideal generated by . Suppose for some , depends on a variable, which is not in . Let denote the ideal generated by , and let denote the ideal generated by and . Then does not contain nonzero elements of . To prove this, suppose that where and . Then , with , and

belongs to the ideal generated by , and so does . This contradicts the choice of since the leading monomial of depends on and is strictly smaller than the leading monomial of . Therefore, the projection of on is dense in , and so, since has dimension , must be zero on some irreducible component of whose projection on is dense in . Since is the Zariski closure of the projection of the -dimensional set , is contained in the Zariski closure of the projection of an irreducible component of . has dimension , hence is zero on , and the projection of on is dense in , which proves that is strongly independent modulo the ideal generated by and .

Here is an example in which needs to be extended. Here , , , and . is zero on one of the two one-dimensional components of :
Extending by results in all depending on only (in fact even constant) while preserving the strong independence of :

Reduce

Reduce can solve arbitrary complex polynomial systems. As the first step, Reduce uses the quantifier elimination algorithm to eliminate the quantifiers. If the obtained quantifier-free system is a disjunction, each term of the disjunction is solved separately, and the solution is given as a disjunction of the solutions of the terms. Thus, the problem is reduced to solving quantifier-free systems of the form

First compute the GroebnerBasis of with variable order and MonomialOrder->Lexicographic, and select the polynomials that do not depend on . Then the solution set of is equal to the solution set of (3) and does not vanish on any component of the zero set of . If contains 1, (3) has no solutions. Otherwise for each , such that the set of elements of depending on and not on any with is not empty, select an element of with the lowest positive degree in . If one of the leading coefficients of is zero on , that is, it belongs to the radical of the ideal generated by , replace by the lexicographic Gröbner basis of the ideal generated by and . Now split the system into

and call the solving procedure recursively on all but the last term of the disjunction (4). Note that the algebraic set is strictly contained in , so the recursion is finite. If the product of all the and belongs to the radical of the ideal generated by , the last term has no solutions. Otherwise, by Property 4, the solution set of the last term is equal to

The conditions guarantee that all the solutions (represented as radicals or Root objects) given by Roots[hij0,xij] are well defined. Reduce performs several operations in order to simplify the inequation conditions returned, like removing multiple factors, removing factors common with earlier inequation conditions, reducing modulo the , and removing factors that are nonzero on .

Options

Options for Reduce, Resolve, and FindInstance

The Wolfram Language functions for solving complex polynomial systems have a number of options that control the way they operate. This section gives a summary of these options.

option name
default value
BacksubstitutionFalsewhether the solutions given by Reduce and Resolve with specified variables should be unwound by backsubstitution
CubicsFalsewhether the Cardano formulas should be used to express solutions of cubics
QuarticsFalsewhether the Cardano formulas should be used to express solutions of quartics

Options of Reduce and Resolve affecting the behavior of complex polynomial systems.

option name
default value
WorkingPrecisionthe working precision to be used in computations, with the default settings of system options; the value of working precision affects only calls to Roots

Options of Reduce, Resolve, and FindInstance affecting the behavior of complex polynomial systems.

Backsubstitution
By default, Reduce may use variables appearing earlier in the variable list to express solutions for variables appearing later in the variable list:
With Backsubstitution->True, Reduce uses backsubstitution to eliminate variables from the right-hand sides of the equations:
Cubics and Quartics
By default Reduce does not use the Cardano formulas for solving cubics or quartics:
Setting the options Cubics and Quartics to True allows Reduce to use the Cardano formulas for solving cubics and quartics:
WorkingPrecision
With WorkingPrecision set to a finite number, Reduce may use numeric methods to find polynomial roots:

The ReduceOptions Group of System Options

Here are the system options from the ReduceOptions group that may affect the behavior of Reduce, Resolve, and FindInstance for complex polynomial systems. The options can be set with

SetSystemOptions["ReduceOptions"->{"option name"->value}].
This sets the option FinitePrecisionGB to True:
This checks the value of FinitePrecisionGB:
This sets the option FinitePrecisionGB back to the default value False:
option name
default value
"AlgebraicNumberOutput"Truewhether Reduce should output AlgebraicNumber objects instead of polynomials in one Root object
"FinitePrecisionGB"Falsewhether finite values of working precision should be used in calls to GroebnerBasis
"ReorderVariables"Automaticwhether Reduce, Resolve, and Solve are allowed to reorder the specified variables
"UseNestedRoots"Automaticwhether Root objects representing algebraic numbers defined by triangular systems of equations can be used in the output

ReduceOptions group options that affect the behavior of Reduce, Resolve, and FindInstance for complex polynomial systems.

AlgebraicNumberOutput

For systems with equational constraints generating a zero-dimensional ideal , the Wolfram Language uses a variant of the CAD algorithm that finds projection polynomials using Gröbner basis methods. If the lexicographic order Gröbner basis of contains linear polynomials with constant coefficients in every variable but the last one (which is true "generically"), then all coordinates of a solution are polynomials in one algebraic number, namely the last coordinate. The setting of AlgebraicNumberOutput determines whether Reduce represents the solution coordinates as AlgebraicNumber objects in the field generated by the last coordinate.

By default, the solution coordinates are represented as AlgebraicNumber objects:
With AlgebraicNumberOutput->False, the solution coordinates are given as polynomials in a Root object:
FinitePrecisionGB
By default, Reduce uses GroebnerBasis with CoefficientDomain->Automatic. This means that even with WorkingPrecision set to a finite number , if the input is exact GroebnerBasis uses exact computations:
Setting the system option "FinitePrecisionGB"->True makes Reduce use GroebnerBasis with CoefficientDomain->InexactNumbers[prec]:
Using finite precision may significantly improve the speed of GroebnerBasis computations. However, the numeric computations may fail due to loss of precision, or give incorrect answers. They usually give less precise results than exact GroebnerBasis computations followed by numeric root finding:
This shows that the results are equal up to their precision:
ReorderVariables
By default, Reduce is not allowed to reorder the specified variables. Variables appearing earlier in the variable list may be used to express solutions for variables appearing later in the variable list, but not vice versa:
Setting the system option "ReorderVariables"->True allows Reduce to pick a variable order that makes the equations easier to solve:
By default, Solve is allowed to reorder the specified variables. In this example it is easier to solve for first and express in terms of the solution for :
Setting the system option ReorderVariables->False makes Solve use the specified variable order:
UseNestedRoots
By default, Reduce, Solve, and FindInstance may express solutions in terms of Root objects representing algebraic numbers defined by triangular systems of equations:
Setting the system option UseNestedRoots->False makes Reduce, Solve, and FindInstance use algebraic numbers defined by univariate equations:
Finding minimal polynomials for all solution coordinates may be slow:
Setting UseNestedRoots->True allows Reduce, Solve, and FindInstance to use algebraic numbers defined by triangular systems that are linear in the last variable:
With the default setting UseNestedRoots->Automatic, Solve represents the second coordinate as a polynomial in the first coordinate:

References

[1] Becker, T. and V. Weispfenning. Gröbner Bases. Springer-Verlag, 1993.
[2] Cox, D., J. Little, and D. O'Shea. Ideals, Varieties, and Algorithms. 2nd ed., Springer-Verlag, 1997.
[3] Łojasiewicz, S. Introduction to Complex Analytic Geometry. Birkhaüser, 1991.