Theorem Hilbert additivity for homogeneous exact sequences
✓ theorem proved

Hilbert additivity for homogeneous exact sequences

Let \(f:M\twoheadrightarrow N\) be a degree-zero surjective linear map of graded modules. Assume the homogeneous pieces of \(M\) are finite-dimensional. Then, for every integer \(n\), \[ \dim M_n=\dim(\ker f)_n+\dim N_n. \] If \(q>0\) and the Hilbert series of \(M\) converges at \(q\), the kernel and target series also converge and \[ H_M(q)=H_{\ker f}(q)+H_N(q). \]

Proof

Step 1. Show surjectivity degree by degree

Given \(y\in N_n\), choose \(m\in M\) with \(f(m)=y\). Decompose \(m=\sum_jm_j\) into its finitely many homogeneous components. Since \(f\) preserves degree, \(y=\sum_j f(m_j)\), with \(f(m_j)\in N_j\). Uniqueness of the direct-sum decomposition gives \(f(m_n)=y\). Thus the restricted map \(f_n:M_n\to N_n\) is surjective.

Step 2. Identify the kernel and apply rank-nullity

An element of \(M_n\) lies in \(\ker f_n\) precisely when its underlying vector is killed by \(f\). This identifies \(\ker f_n\) with \((\ker f)\cap M_n\); the identification is the identity on underlying vectors in both directions.

The image \(N_n\) of the finite-dimensional space \(M_n\) is finite-dimensional. Rank-nullity for \(f_n\), followed by this kernel identification, proves the dimension formula.

Step 3. Check convergence of both summands

The two dimensions on the right are nonnegative and individually at most \(\dim M_n\). Multiplying by \(q^n>0\) preserves these bounds. Each series is therefore dominated termwise by the convergent nonnegative series for \(M\), which proves convergence of both.

Step 4. Sum the exact identities

Multiply the degreewise dimension equation by \(q^n\) and sum over \(n\in\mathbb Z\). Step 3 permits splitting the sum into the two convergent series, yielding the stated identity. The equality is obtained from finite-dimensional rank–nullity in each piece and summability of the two resulting nonnegative series.

References

The precise Lean sources are listed below.

Lean formalization

OAI.KLInvariance.Hilbert.degreeMap_surjective: HilbertSeries.lean:202.

OAI.KLInvariance.Hilbert.finrank_exact: HilbertSeries.lean:360.

OAI.KLInvariance.Hilbert.value_exact_of_summable: HilbertSeries.lean:392.

OAI.KLInvariance.Hilbert.degreeKernelEquiv: HilbertSeries.lean:346.

OAI.KLInvariance.Hilbert.value_exact: HilbertSeries.lean:374.

Used by