Back to search
2202.09501

The Adaptive Spectral Koopman Method for Dynamical Systems

Bian Li, Yi-An Ma, J. Nathan Kutz, Xiu Yang

correctmedium confidence
Category
math.DS
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper’s Theorem 3.2 shows that ASK solutions are real-valued when x, f, g (hence K) are real by pairing each complex eigen-contribution with its conjugate in the truncated reconstruction g_N(x(t)), explicitly using the middle entry of eigenvectors and a direct algebraic simplification to a real expression (Section 3.5; Lemma 3.1 and Theorem 3.2, and eqs. (14)–(16) for setup) . The candidate model solution proves the same claim via the conjugate-pair structure for real K, uniqueness of the Vc=g(Ξ) expansion with real right-hand side, and an explicit observation that the ASK amplitude a_j=c_j(v_j)_{mid} is invariant under eigenvector rescaling, so conjugate pairs contribute 2 Re(a_j e^{λ_j t}). Both arguments rely on the same core facts and reach the same conclusion; the model adds clarity on coefficient conjugacy and rescaling invariance, while the paper uses a concrete inner-product argument and explicit algebra to show the sum is real (see the paper’s computation of cm ν e^{λ t} + c̄m ν̄ e^{λ̄ t} yielding a real expression) .

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The paper’s proof of real-valuedness is correct and well-motivated within the ASK framework, relying on standard linear algebra and the conjugate-pair structure of real matrices. The exposition would benefit from making implicit assumptions explicit (diagonalizability/invertibility of V) and from a brief remark on normalization invariance of the amplitude and the need to include full conjugate pairs in any truncation. These small clarifications would strengthen the presentation for both theoreticians and practitioners.