The Quadratic Sieve: Linear Algebra Stage
Finding dependencies in the relation matrix over the field with two elements, and why this stage is the practical bottleneck.
Engineering Mathematics articles in the KEVOS Engineering library. 1073 pages.
Finding dependencies in the relation matrix over the field with two elements, and why this stage is the practical bottleneck.
The sieving stage: identifying smooth polynomial values in bulk using logarithm accumulation rather than trial division.
The regulator as the covolume of the unit lattice, its computation, and the precision and verification it demands.
Constructing resolvent polynomials whose factorisation distinguishes candidate Galois groups, and the practical issues in using them.
The Round 2 algorithm: computing the maximal order prime by prime via radicals and rings of multipliers.
The algorithm that drives any matrix to reduced row-echelon form: pivot search, interchange, normalisation, column clearing, cost and pivoting strategy.
RSA key generation, encryption, decryption and correctness, together with the assumptions its security depends on.
Factoring via class groups of quadratic orders, and its place as the conceptual bridge to the elliptic curve method.
The classical sieve for enumerating primes, its complexity, segmented variants, and its role as a precomputation step.
The Smith normal form, its computation by alternating row and column reduction, and the invariant factors it exposes.
Linear combinations and spans in an abstract vector space: the definition, the proof that a span is always a subspace, and how membership becomes a linear system.
Representing field elements as coefficient vectors relative to a power basis or integral basis, with a common denominator.
The structure theorem decomposing every finite abelian group into cyclic factors, and its computational consequences.
The structure of Z_n* as a product of cyclic groups, the Carmichael function, and why prime moduli behave differently.
Practical considerations in running class group computations: parameter tuning, parallelism, precision management and diagnostics.
The sub-resultant remainder sequence: predicting the divisible factor at each step to keep coefficients near minimal without content computation.
Finding the subfields of a number field, by lattice methods and by linear algebra over the complex numbers.
Trace, norm and characteristic polynomial of a field element, their computation, and their use as invariants and cross-checks.
Trial division as a primality test and as a filter, its exponential cost, and the role it still plays in practice.
Trial division as the first factoring step, its cost, and Lehman's improvement on Fermat's method.
Choosing the trial division bound ahead of a probabilistic test, and the cost balance that determines it.
Content and primitive part, Gauss's lemma, and why factoring over the rationals reduces to factoring over the integers.
Unique factorisation domains, the distinction between irreducible and prime, and the standard examples and counterexamples.
Why Euclidean domains are principal ideal domains and why principal ideal domains have unique factorisation.
Unique factorisation in polynomial rings over a field, and the extension to polynomial rings over a UFD.
The fundamental theorem of arithmetic: existence and uniqueness of prime factorisation, and why the uniqueness half is the difficult one.
Proof that a matrix has exactly one reduced row-echelon form: pivot columns agree by induction, ranks agree, and rows are forced to coincide entry by entry.
The analytic inequalities, series estimates and elementary bounds relied on repeatedly in the analysis of number-theoretic algorithms.
Valuations at prime ideals, uniformising elements, and computing the exponent of a prime in an ideal factorisation.
Express every solution of a linear system as a fixed vector plus a linear combination of n-r vectors read directly from the reduced row-echelon form.
Vector representation relative to an ordered basis: the coordinate map is a well-defined, injective and surjective linear transformation onto complex n-space.
The ten defining properties of a vector space: closure, commutativity, associativity, zero vector, additive inverses, distributivity and the unit scalar.
The ten algebraic properties of column vector addition and scalar multiplication, how each is proved entrywise, and why they license later manipulation.
The ten vector space properties of matrix addition and scalar multiplication: closure, commutativity, associativity, the zero matrix, additive inverses and distributivity.
Vector spaces over a field, the well-definedness of dimension, and the rank-nullity relation.
Confirming class group and regulator results against the analytic class number formula, and what such confirmation does and does not establish.
General and short Weierstrass forms, the discriminant and j-invariant, and the transformations relating equivalent models.
What makes an equation linear, why flatness matters, and how addition and scalar multiplication alone generate the whole of linear algebra in any dimension.
A gallery of computed spectra: distinct, repeated, defective, complex and zero eigenvalues, with algebraic and geometric multiplicities for each eigenspace.
Finitely generated abelian groups as integer matrix problems, and the two normal forms that answer the two basic questions about them.
Zero divisors, integral domains, and why the absence of zero divisors is what makes cancellation and root counting work.
Counting points over finite fields, the Hasse bound, and how local counts assemble into a global zeta function.
Additive and abelian categories, the axioms, exactness in a general abelian category, the Freyd-Mitchell embedding theorem, and the standard examples.
Abelian groups, subgroups, cosets and quotient groups, Lagrange's theorem, homomorphisms and isomorphism theorems, cyclic groups and the structure theorem for finite abelian gro…
Adjoint pairs, unit and counit, the tensor-hom adjunction, preservation of limits and colimits, and the exactness consequences that make adjointness central to homological algebra.
Algebraic numbers and integers, minimal polynomials, number fields as finite extensions of Q, real and complex embeddings, the signature, and the primitive element theorem.
Algebras of arbitrary type: signatures, arities, the significance of nullary operations, and how the choice of type determines subalgebras, homomorphisms and the whole subsequen…
Computing minimal Weierstrass models, Tate's algorithm for reduction type and conductor, torsion subgroup determination, heights, and descent for rank computation.