Sylvester matrix

Sylvester matrix
Name	Sylvester matrix
Field	Linear algebra, Algebraic geometry, Computational algebra
Introduced	19th century
Inventor	James Joseph Sylvester

Sylvester matrix The Sylvester matrix is a structured matrix associated with two polynomials used to study common roots and elimination problems. It connects classical work by James Joseph Sylvester, nineteenth-century developments at University of Cambridge, and later computational approaches from institutions like Massachusetts Institute of Technology and École Polytechnique. The construction underpins links between elimination theory in Galois theory, resultant computations in David Hilbert's program, and modern algorithms implemented at organizations such as INRIA and Los Alamos National Laboratory.

Definition

The Sylvester matrix of two univariate polynomials is defined using the coefficients of two polynomials to form a block-structured matrix whose determinant gives the resultant. The concept originates with James Joseph Sylvester and complements work by Arthur Cayley, Évariste Galois, and Niels Henrik Abel. It is a central object in the study of polynomial factorization pursued by scholars at University of Göttingen, University of Oxford, and University of Paris. Applications and theoretical developments were advanced by researchers affiliated with Princeton University, University of Chicago, and University of Cambridge.

Construction and properties

The Sylvester matrix is constructed by arranging coefficient vectors of two polynomials into a Toeplitz-like block matrix; its size relates to the degrees of the polynomials. Key algebraic properties were elucidated in correspondence among James Joseph Sylvester, Arthur Cayley, and contemporaries active in Royal Society. Determinantal properties connect to work by Carl Friedrich Gauss on roots and by Leopold Kronecker on elimination. The matrix exhibits rank conditions that detect common factors, paralleling results in papers from University of California, Berkeley and seminars at Institut des Hautes Études Scientifiques. Over principal ideal domains studied at University of Bonn and ETH Zurich, Smith normal form analyses of the Sylvester matrix link to invariant factor theories developed by Emmy Noether and Richard Dedekind.

Relation to resultant and discriminant

The determinant of the Sylvester matrix equals the resultant of two polynomials, a classical invariant explored by David Hilbert and used in proofs by Hilbert and Emil Artin. Resultant computations via Sylvester matrices are foundational in elimination theory, informing methods in Alexander Grothendieck's algebraic geometry school at Institut des Hautes Études Scientifiques and later work by Jean-Pierre Serre. The discriminant of a polynomial emerges from the resultant of the polynomial and its derivative; this link was central for studies of singularities by Bernhard Riemann and analytic continuation considerations by Henri Poincaré. Connections to projective geometry and intersection theory reflect themes from Oscar Zariski and Federico Bézout's antecedent theorems.

Computational methods and algorithms

Algorithms for forming and manipulating Sylvester matrices appear in computer algebra systems developed at Symbolics, Wolfram Research, and SageMath. Numerical stability concerns prompt use of structured linear algebra techniques from researchers at IBM Research and Google DeepMind who adapted singular value decomposition and rank-revealing QR factorizations. Efficient resultant computation strategies draw on subresultant PRS algorithms by Joseph Kronecker descendants and modular algorithms popularized by groups at Max Planck Institute for Mathematics and University of Waterloo. Polynomial GCD computations using Sylvester matrices are implemented in libraries from Microsoft Research and integrated into packages at Cornell University and Stanford University.

Applications

Sylvester matrices are used in elimination theory for solving polynomial systems, as applied in robotics research at Carnegie Mellon University and control theory studies at California Institute of Technology. They appear in coding theory work at Bell Labs and cryptanalysis investigations by teams at National Security Agency and academic groups at École Normale Supérieure. In computational algebraic geometry, Sylvester matrices support computation of resultants for curve intersection problems tackled by scholars at University of Illinois Urbana-Champaign and University of Texas at Austin. Applications in signal processing relate to research at Massachusetts Institute of Technology and Stanford University, while computer vision problems benefiting from elimination techniques have been pursued at University of Toronto and University College London.

Examples and special cases

For two quadratics, the Sylvester matrix becomes a 4×4 matrix whose determinant yields a degree-4 resultant; classical treatments appear in textbooks by G. H. Hardy and John Edensor Littlewood. For one linear polynomial and one cubic, the resulting structured matrix reduces to a 4×4 Toeplitz-like form studied in lectures at Imperial College London. The concept generalizes to multivariate resultants via Macaulay matrices developed by F.S. Macaulay and extended by researchers at University of Leiden and University of Sydney. Special cases include the Bézout matrix and Hankel-structured variants examined by analysts at Yale University and Princeton University.

Category:Linear algebra