KEVOS
ArticlesServicesCase studiesAboutContact
ArticlesServicesCase studiesAboutContact
← ArticlesTest Polynomials for Galois Group SoftwareEngineering · Engineering MathematicsLesson 825/884← PrevNext →
ArticlePublished 7 Aug 20262 min readBy Kevin Jogintest polynomialsverificationsoftware testingGalois group
On this page

Ask about this page

KEVOS AITest Polynomials for Galois Group Software

KEVOS knowledge first · trusted web sources when needed

Galois Groups and Field Families

Test Polynomials for Galois Group Software

Why Galois group implementations need a curated test set, what a good set covers, and how to build one.

Engineering / MathematicsGalois Groups and Field Families2 min readKV-MATH-0623

Galois group computation is error-prone and its errors are quiet: a misidentified group is a plausible-looking answer. A curated set of polynomials with known groups is the only practical defence.

What a good test set covers

Coverage requirements for a Galois group test set
CategoryWhy it matters
One polynomial per transitive group, each degreeBasic coverage; catches gross misidentification
Polynomials whose resolvents have repeated rootsExercises the Tschirnhaus fallback path
Large coefficientsExercises coefficient growth handling
Small discriminantsFields with many ramified primes
Imprimitive groups in composite degreeExercises subfield detection
Near-miss pairsGroups distinguished only by a specific resolvent

Key point

The near-miss pairs are the most valuable entries. Any group pair that a single invariant fails to separate is exactly where an implementation will silently return the wrong answer, and generic random testing will rarely produce such a case.

Constructing test polynomials

Building test cases with known groups

  1. Start from a known groupConstruct a field with the desired group, for instance as a subfield of a cyclotomic field.
  2. Find a defining polynomialCompute the minimal polynomial of a generator.
  3. ReduceApply polynomial reduction to obtain small coefficients.
  4. RecordStore the polynomial with its verified group, discriminant and any subfields.

Note

Abelian groups are easy to realise inside cyclotomic fields by the Kronecker-Weber theorem. Non-abelian groups require more work, which is why published test sets are valuable — they encode considerable effort.

Cross-checking

Beyond a fixed test set, results should satisfy structural constraints, and checking these catches errors on inputs not in the set.

  • The group order must be divisible by the field degree, since the group is transitive.
  • The group must lie in the alternating group exactly when the discriminant is a square.
  • Observed cycle types must all occur in the identified group.
  • The subfield lattice must match the subgroup lattice.

Key point

The cycle type check is the strongest cheap invariant. Any factorisation shape observed modulo an unramified prime must correspond to an element of the identified group, and a single violation disproves the identification outright.

Regression discipline

Cost

Galois group code accumulates special cases. A regression suite run on every change is the only way to keep those cases correct, and the cost of running it is trivial compared to the cost of shipping a silent misidentification — see implementation pitfalls and testing.

Source. Henri Cohen, A Course in Computational Algebraic Number Theory, Springer GTM 138 — 6.3.7. Structural reference unverified: the source file was not available during authoring; chapter and section numbers are taken from the published edition and have not been checked against a physical copy.

Related pages

  • Galois Groups of Sextic and Septic Fields
  • Number Theory Software Packages
  • Implementation Pitfalls and Testing Strategy
  • Constructing Tables of Number Fields

Continue learning

Galois Groups of Sextic and Septic FieldsArticle · Engineering MathematicsNEXT LESSON →Constructing Tables of Number FieldsArticle · Engineering MathematicsGalois Groups of Quintic FieldsArticle · Engineering MathematicsCyclic and Pure Cubic Field FamiliesArticle · Engineering Mathematics
KEVOS · Engineering, manufacturing and project improvement
ArticlesServicesCase studiesAboutContact
© 2026 KEVOS®