The conjugation action of the special linear group on matrices #
The special linear group SL(n, R) acts on the square matrices Matrix n n R by conjugation,
g • M = g M g⁻¹. As for any conjugation action, it is recorded as an action of the type synonym
ConjAct (SL(n, R)), so that it does not clash with the action of SL(n, R) on column vectors
or with the scalar action of R. Mathlib provides the conjugation action ConjAct.unitsScalar
of the units Mˣ of a monoid on M, which covers GL(n, R) acting on Matrix n n R, but not
the action of SL(n, R), whose orbits are the SL(n, R)-conjugacy classes of matrices.
Conjugation preserves the trace, the determinant and, since adjugate g = g⁻¹ for g ∈ SL(n, R),
commutes with taking adjugates.
Conjugation by SL(n, R) is conjugation by its image in GL(n, R), so SL(n, R)-conjugates are
GL(n, R)-conjugates. Conversely, a conjugate g X g⁻¹ by g ∈ GL(n, R) is also a conjugate by
SL(n, R) exactly when the determinant of g can be corrected inside the centralizer of X: some
invertible matrix commuting with X has the same determinant as g.
Main definitions #
ConjAct.specialLinearGroupMulAction: the conjugation actiong • M = g M g⁻¹ofConjAct (SpecialLinearGroup n R)onMatrix n n R, with scalar multiplicationConjAct.specialLinearGroupScalar.TauCeti.traceDetFiber n t d: then × nmatrices of tracetand determinantd(TauCeti.mem_traceDetFiber), as a sub-action, whose orbits are theSL(n, R)-conjugacy classes of such matrices.
Main results #
ConjAct.specialLinearGroup_smul_def:g • M = g M g⁻¹.ConjAct.trace_specialLinearGroup_smulandConjAct.det_specialLinearGroup_smul: conjugation preserves the trace and the determinant.Matrix.SpecialLinearGroup.trace_mul_mul_inv: the product formtrace (P k P⁻¹) = trace kof trace invariance, forP k : SL(n, R).ConjAct.adjugate_specialLinearGroup_smul: conjugation commutes with the adjugate.ConjAct.toConjAct_toGL_smul: conjugating bys ∈ SL(n, R)is conjugating by its image inGL(n, R).Matrix.GeneralLinearGroup.toConjAct_smul_mem_orbit_specialLinearGroup_iff: aGL(n, R)-conjugate is anSL(n, R)-conjugate exactly when the determinant of the conjugator is the determinant of an invertible matrix in the centralizer.Matrix.GeneralLinearGroup.orbit_conjAct_specialLinearGroup_subset: anSL(n, R)-conjugate is aGL(n, R)-conjugate.TauCeti.adjugate_mem_traceDetFiber: for2 × 2matrices, the adjugate has the same trace and determinant.TauCeti.traceDetFiberAddScalarEquiv: for2 × 2matrices, translationM ↦ M + rby a scalar matrix maps the matrices of tracetand determinantdbijectively to those of tracet + 2 rand determinantd + r t + r².
References #
- A. Popa and D. Zagier, An elementary proof of the Eichler–Selberg trace formula,
J. Reine Angew. Math. 762 (2020), 105–122, arXiv:1711.00327, §1, equation (2): the sum
over the
SL(2, ℤ)-conjugacy classes of integer matrices of fixed trace and determinant, whose orbits this action describes.
SL(n, R) acts on Matrix n n R by conjugation, g • M = g M g⁻¹, through the type synonym
ConjAct, in the manner of Mathlib's ConjAct.unitsScalar.
Conjugation g • M = g M g⁻¹ is an action of SL(n, R) on Matrix n n R.
Conjugation by SL(n, R) preserves the trace.
Conjugation by SL(n, R) preserves the determinant.
Conjugation by SL(n, R) commutes with the adjugate:
adjugate (g M g⁻¹) = g (adjugate M) g⁻¹.
Conjugating a matrix by the image of s ∈ SL(n, R) in GL(n, R) is conjugating it by s.
Determinant correction of conjugators. Over a commutative ring R, the conjugate
g X g⁻¹ of a matrix X by g ∈ GL(n, R) is also a conjugate of X by SL(n, R) exactly when
some invertible matrix commuting with X has the determinant of g.
An SL(n, R)-conjugate of a matrix is a GL(n, R)-conjugate.
Conjugation in SL(n, R) preserves the trace: trace (P k P⁻¹) = trace k. This is the
product form of ConjAct.trace_specialLinearGroup_smul.
The matrices of trace t and determinant d, as a sub-action of the conjugation action of
SL(n, R): its orbits are the SL(n, R)-conjugacy classes of such matrices.
Instances For
A matrix lies in traceDetFiber n t d exactly when its trace is t and its determinant d.
A 2 × 2 matrix lies in traceDetFiber (Fin 2) t d exactly when its entries satisfy
M₀₀ + M₁₁ = t and M₀₀ M₁₁ - M₀₁ M₁₀ = d.
The adjugate of a 2 × 2 matrix M has the trace and the determinant of M: it lies in
traceDetFiber (Fin 2) t d whenever M does.
Adding the scalar matrix r to a 2 × 2 matrix of trace t and determinant d gives a matrix
of trace t + 2 r and determinant d + r t + r².
Translation M ↦ M + r by a scalar matrix, from the 2 × 2 matrices of trace t and
determinant d to those of trace t + 2 r and determinant d + r t + r².
Instances For
traceDetFiberAddScalarEquiv t d r adds the scalar matrix r.
The inverse of traceDetFiberAddScalarEquiv t d r subtracts the scalar matrix r.