Back to search
2506.18352

Brown–Voiculescu entropy revisited

Bhishan Jacelon, Robert Neagu

correcthigh confidence
Category
Not specified
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper proves Theorem 5.4 rigorously: for simple, separable, unital, nuclear, Z-stable C*-algebras with finitely many extremal traces and UCT, not AF, one has htdr(α) = sup_{τ∈∂eT(A)} hqd_τ(α) = sup_{τ∈T(A)} hqd_τ(α). The proof first shows htdr ≥ hqd_τ for every trace (Proposition 5.3), then constructs 2-colour, contractive decomposable approximations from quasidiagonal systems of the finitely many extremal traces, yielding htdr ≤ max_i hqd_{τ_i} and hence the equality (Theorem 5.4, with the key steps (5.1)–(5.7)) . By contrast, the model solution conflates nuclear dimension with decomposition rank in step (i), incorrectly claiming CETWW’s dim_nuc(A)=1 gives contractive two-colour (decomposition rank) approximations; contractivity is part of decomposition rank, not nuclear dimension. It also incorrectly asserts that in the AF case both sides are zero, contradicting the paper’s AF computations where htdr equals Voiculescu’s approximation entropy and can be positive (e.g., log k for a UHF Bernoulli shift) .

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The variational-type principle relating contractive coloured entropy to quasidiagonal entropies of extremal traces is clean and well motivated. The proof leverages modern classification and Z-stability with careful rank bookkeeping to avoid pitfalls (e.g., reindexing that alters finite-dimensional sizes). The exposition is clear, with helpful remarks explaining the necessity of the finiteness of the tracial boundary and how to avoid the “not AF” hypothesis via a modified definition. Minor clarifications would further aid readers in tracking the role of decomposition rank versus nuclear dimension.