A solution to the inverse generator problem and related questions
Emiel Lorist, Martin Meyries, Mark Veraar
math.FA
Aug 6, 2026 · v3
math.NA
TL;DR
The proof of the main finite-dimensional matrix construction (Theorem 1.1) was formalized in Lean 4, with AI assistance, and released as an artifact.
Abstract
We give a negative solution to the inverse generator problem on Hilbert spaces. More precisely, we construct a bounded operator $A$ with dense range on a Hilbert space $H$ that generates a bounded, strongly stable $C_0$-semigroup, while $A^{-1}$ does not generate a $C_0$-semigroup. We also construct an exponentially stable generator $A$ with $0 \in ρ(A)$ such that the inverse semigroup is unbounded and grows at least double logarithmically. For the latter generator, every Cayley transform satisfies the ordinary Kreiss resolvent condition but is neither strongly Kreiss bounded nor power bounded. Moreover, its powers satisfy a doubly logarithmic lower bound. Therefore, the Crank–Nicolson scheme is unstable in operator norm both for every fixed step size over long times and under mesh refinement at any fixed final time. Our counterexamples are deduced from a common finite-dimensional construction. For $α\in(0,1)$, we use explicit bases of $\mathbb C^{2n}$ whose partial-sum projections are uniformly bounded and whose unconditionality constants are comparable to $n^α$. The matrices underlying the counterexamples are then obtained as Schauder multipliers with respect to these bases, using a sequence of eigenvalues whose moduli decay doubly exponentially.
Problem
The inverse generator problem asks whether, on a Hilbert space, A^{-1} generates a C_0-semigroup whenever A generates a bounded C_0-semigroup and has dense range. Related open questions concern power boundedness and strong Kreiss boundedness of Cayley transforms, and the stability of the Crank–Nicolson scheme.
Approach
A finite-dimensional theorem gives explicit invertible, diagonalizable matrices A_n in M_{2n}(C) with uniformly bounded exponentials, while ||e^{πA_n^{-1}}|| is comparable to n^α. The matrices are Schauder multipliers with respect to explicit bases. These bases have uniformly bounded partial-sum projections and unconditionality constants of order n^α, and the eigenvalue moduli decay doubly exponentially. Infinite-dimensional counterexamples are built as direct sums of these matrices, and the proof of the finite-dimensional theorem was formalized in Lean 4.
Results
The inverse generator problem on Hilbert spaces has a negative answer. An exponentially stable generator is constructed whose inverse semigroup grows at least double logarithmically. Its Cayley transforms satisfy the Kreiss condition but are neither strongly Kreiss bounded nor power bounded, so Crank–Nicolson is unstable in operator norm.