{"id":"2c8c13c4-153e-4f91-9cf7-72275b92f54f","arxiv_id":"1908.08824","paper_version":6,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A square matrix is of maximal trace over rotations exactly when it is symmetric and has at most one negative eigenvalue whose absolute value does not exceed the others.","lead":"This paper characterizes which matrices give the maximum trace when multiplied by any rotation, in terms of symmetry and eigenvalues. It also offers rotation computations that avoid the singular value decomposition for 2x2 and 3x3 cases.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The only real defect is a garbled determinant identity in Prop. 3; it is a repairable typo and does not threaten Corollary 1.","rationale":"The reader's ACCEPT verdict is justified. Corollary 1 is the central claim, and its proof is structurally sound: Proposition 2 correctly forces symmetry via a Givens rotation; Proposition 5 gives the sufficiency direction using the refined bound of Proposition 4; Proposition 6 gives necessity by exhaustive analysis of the determinant sign and the number and magnitude of negative eigenvalues. The only substantive flaw is the garbled determinant identity inside Proposition 3(3). That line is wrong as printed, but the needed fact, that an improper orthogonal matrix has eigenvalue -1, is standard and is immediately recovered from the corrected identity det(I+W) = det(W)det(I+W^T). All subsequent steps of Proposition 3(3) are algebraically valid once that fact is in place, so the sufficiency argument survives. The secondary algorithms for d = 2, 3 are not central to the main theorem, and the paper is explicit about Newton's method failing on rank-one inputs and angle-pi rotations, with an SVD fallback. The reported timing comparison is a benchmark claim and does not bear on correctness of the characterization. Thus no verdict change is needed; the paper should be accepted with the determinant line corrected or annotated.","tokens_in":25240,"tokens_out":10740,"duration_ms":108376,"concrete_test":"Re-derive the opening line of Proposition 3(3) as det(I+W) = det(W)det(I+W^T) and verify that for det(W) = -1 this implies det(I+W) = 0; then check that the remainder of the block proof, yielding Wbar = [[W0,O],[O^T,-1]] and tr(WD) <= sum_{j=1}^{d-1} sigma_j - sigma_d, follows without further changes. If the corrected identity supports the same conclusion, the central characterization is unaffected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step for the sufficiency direction of Corollary 1 is Proposition 3(3), used through Propositions 4 and 5(2). The proof must show that an orthogonal W with det(W) = -1 has eigenvalue -1. The printed chain, det(W+I) = -det(W)det(W+I) = -det(W^T)det(W+I) = -det(I+W^T) = -det(I+W), contains a false equality: det(W^T)det(W+I) is not generally det(I+W^T). The correct identity is det(I+W) = det(W)det(I+W^T) = det(W)det(I+W), which, for det(W) = -1, gives det(I+W) = -det(I+W), hence det(I+W) = 0. With that correction, the subsequent block decomposition Wbar = [[W0,O],[O^T,-1]] and the inequalities tr(W0S0) <= tr(S0) and gamma >= sigma_d go through as written, so Proposition 3(3), Proposition 4, and Proposition 5(2) are sound. This is a typo-level defect in a standard supporting fact, not a gap in the characterization itself. The necessity direction in Proposition 6 is exhaustive and correct, and the secondary numerical material is honestly caveated. No other load-bearing assumption is left unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the problem of maximizing tr(UM) over rotation matrices U for a fixed real d x d matrix M. Its main result (Corollary 1) characterizes matrices of maximal trace over rotations: after symmetry is established, at most one eigenvalue may be negative, and if one is negative, its absolute value must not exceed that of any other eigenvalue. The proof is developed in Section 3, with sufficiency proved through Propositions 3-5 and necessity through Proposition 6. For d=2 and d=3, the characterization is turned into simple tests (nonnegative trace for d=2; positive semidefiniteness of tr(A)I-A for d=3), and the paper then proposes SVD-free computational procedures: closed-form two-dimensional formulas, a trigonometric eigensolver for symmetric 3x3 matrices, and a Cayley-transform/Newton method for nonsymmetric 3x3 matrices with an SVD fallback. The numerical procedures are implemented in Fortran and Matlab and compared in Table 1.","tokens_in":25492,"tokens_out":15664,"duration_ms":159728,"significance":"If correct, the main characterization is clean and practically useful: checking whether the identity rotation is optimal reduces to a symmetry test and an eigenvalue count. The sufficiency proof is self-contained and does not rely on the SVD, and the necessity argument exhaustively handles the determinant sign cases. The paper also provides reproducible code and honestly describes the failure modes of the Newton-based method, which strengthens the numerical portion. The main defect I found is a local typographical error in the determinant argument of Proposition 3; it does not affect the validity of the theorem.","major_comments":[],"minor_comments":[{"comment":"The displayed determinant chain contains a false equality: det(W^T)det(W+I) is not generally equal to det(I+W^T). The correct identity to invoke is det(I+W) = det(W)det(I+W^T) = det(W)det(I+W), which, together with det(W)=-1, gives det(I+W)=0. Once this line is corrected, the rest of the proof of Proposition 3(3) goes through unchanged.","section":"Section 3, Proposition 3(3)"},{"comment":"The assertion that a matrix of maximal trace over rotation matrices has nonnegative trace needs the standing assumption d >= 2. For d=1, the matrix [-1] is maximal over the only rotation matrix [1] but has trace -1. Please state the dimension assumption explicitly; the proof of Proposition 2 already requires d >= 2.","section":"Corollary 1, last sentence"},{"comment":"There is a typographical error in the chain tr(WD) = tr(W_hat D_hat) =<= ...; the symbol '=<' should be '=' or the words should be split into two inequalities.","section":"Proposition 4"},{"comment":"The claim that Newton's method is successful 'close to one hundred percent' of the time, and the statement that rank-one matrices are not successfully handled, are not quantified. Please report the number of failures and the rank breakdown, since Table 1 reports a different experiment (one million random matrices, all of which are stated to have been solved).","section":"Section 6, numerical experiments"}],"recommendation":"minor_revision","confidential_remarks":"The only mathematical flaw is the determinant typo in Proposition 3; it is local and easily corrected. I have no concerns about novelty or citation behavior beyond the minor suggestions in the report."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe main result is exactly what the abstract promises: a d×d matrix has maximal trace over rotations iff it is symmetric and has at most one negative eigenvalue, whose absolute value does not exceed the other eigenvalues. The proof is elementary and mostly self-contained; the necessity is split by determinant sign, and the sufficiency reduces to the nonnegative diagonal case via a conjugation argument. The orthogonal version is credited to Horn and Johnson; the rotation version is, as far as I can tell, genuinely new in this explicit form.\n\nWhat the paper does well: the characterization is clean and checkable. For d=2 it collapses to trace nonnegativity; for d=3 to positive semidefiniteness of tr(A)I − A. The closed-form eigensolver for 3×3 symmetric matrices is a solid piece of classical linear algebra, and the Newton–Cayley procedure for nonsymmetric M is honestly documented: the authors report failure modes (rank-one inputs, π rotations), provide an SVD fallback, publish Fortran and Matlab code, and report timings that show no speed advantage over the SVD in Fortran. That honesty earns credit.\n\nI was told to look at a garbled determinant identity in Proposition 3(3). On inspection the printed chain is correct: det(W^T)det(W+I) = det(W^T(W+I)) = det(I+W^T), so the conclusion det(I+W)=0 stands. The argument is a bit roundabout but valid.\n\nSoft spots, in proportion: the computational contribution is secondary, and the paper says so. The 2D closed forms are standard, and the 3D eigensolver builds on known references. The Newton method's convergence is heuristic, but the authors don't oversell it. Minor typos: a '=≤' slip in Proposition 4's proof, and the prose could be tighter. Nothing affects the mathematics. The only self-citation is [9], restated with full proof, so the citation pattern is fine.\n\nWho it's for: anyone working on Procrustes or Wahba problems, especially in small dimensions, and anyone who wants an eigenvalue characterization of rotation optimality. It deserves a serious referee; I would accept.","headline":"A clean, correct characterization of maximal-trace rotations; the alleged proof flaw doesn't survive contact with the paper.","tokens_in":26002,"tokens_out":8567,"would_cite":true,"duration_ms":74237,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["15A18","15A42","65H17","65K99","93B60"],"pacs":[],"model":"deepseek-v4-flash","headline":"A $d\\times d$ matrix is of maximal trace over rotation matrices if and only if it is symmetric and has at most one negative eigenvalue, whose absolute value does not exceed the other eigenvalues.","keywords":["eigenvalues","orthogonal","Procrustes","rotation","singular value decomposition","trace","Wahba","maximal trace"],"falsifier":"Take $A=\\operatorname{diag}(1,-2)$. The paper's condition says $A$ should not be maximal over rotations; indeed, rotating by $\\pi$ gives $\\operatorname{tr}(UA)=1>-1=\\operatorname{tr}(A)$. To challenge the opposite direction, take $A=\\operatorname{diag}(1,-1/2)$ and check numerically over a fine grid of rotation angles that no rotation exceeds $\\operatorname{tr}(A)=1/2$. A decisive scan would randomly generate symmetric matrices, compute their eigenvalue conditions, and compare with a numerical maximization of $\\operatorname{tr}(UA)$ over $\\operatorname{SO}(d)$.","tokens_in":25032,"feed_emoji":"🔄","tokens_out":10174,"duration_ms":99311,"temperature":0.7,"pith_summary":"This paper establishes when a given real matrix already attains the maximum possible trace among all of its left multiplications by rotation matrices—equivalently, when the identity rotation is the optimal alignment in the constrained orthogonal Procrustes problem and in Wahba's weighted version. The characterization it proves is compact: a $d\\times d$ matrix is maximal over rotations if and only if it is symmetric and has at most one negative eigenvalue, whose absolute value, when present, is no larger than the other eigenvalues. This turns what was a global optimization question into a symmetry test and an eigenvalue sign count. For dimensions two and three the paper demonstrates how to use the characterization directly, and it offers SVD-free algorithms for producing the optimal rotation when the identity is not already optimal.","feed_headline":"Rotation-optimal matrices are symmetric and almost positive semidefinite","feed_subtitle":"Checking whether identity already solves a point alignment reduces to symmetry and a sign count.","key_machinery":"The load-bearing mechanism is a trace bound for diagonal matrices under determinant-minus-one orthogonal transformations: if $D$ is diagonal with nonnegative entries and $W$ is orthogonal with $\\det(W)=-1$, then $\\operatorname{tr}(WD)$ is at most the sum of all diagonal entries except the smallest, minus that smallest entry. This bound, proved as Proposition 3 and sharpened in Proposition 4, lets the sufficiency argument reduce rotation maximality to a spectrum sign count. The necessity direction uses a Givens rotation construction: whenever a matrix is not symmetric, a carefully chosen plane rotation raises the trace, so maximality forces symmetry. For $d=3$ the resulting practical test is the positive semidefiniteness of $S=\\operatorname{tr}(A)I-A$.","core_discovery":"The paper's central claim is a necessary-and-sufficient eigenvalue characterization of maximal trace over rotations: a real $d\\times d$ matrix $A$ satisfies $\\operatorname{tr}(UA)\\leq \\operatorname{tr}(A)$ for every rotation $U$ exactly when $A$ is symmetric and its eigenvalues contain at most one negative number, that number being no larger in absolute value than the others. The proof forces symmetry by exhibiting a Givens rotation that strictly increases the trace of any nonsymmetric matrix, then shows by an eigenvalue sign-flipping argument that a maximizing matrix cannot have two negative eigenvalues or one oversized negative eigenvalue. Conversely, a spectral bound on orthogonal matrices of determinant minus one shows that no rotation can improve the trace once the symmetric matrix satisfies the eigenvalue condition. The authors state the result as Corollary 1 and treat it as the main goal of the paper; the SVD-free computational recipes in later sections are a secondary contribution.","pith_inferences":["Because the $d=3$ test is a linear matrix inequality, membership in the set of rotation-maximal matrices is a semidefinite programming feasibility problem; the paper does not draw this connection, but it means existing SDP solvers could certify optimality in the same step that checks the condition.","The two-dimensional closed form depends only on two scalar sums (a dot-product sum and a signed-area sum), so weighted 2D Procrustes reduces to computing those two numbers and normalizing a vector; this is a simpler pipeline than calling a general SVD routine.","The eigenvalue characterization is a symmetry-plus-spectrum condition, which suggests that the same 'almost positive semidefinite' notion may appear in other optimization problems over compact Lie groups where the objective is a trace; exploring that analogy would go beyond this paper.","For odd dimensions the paper proves the semidefinite test is necessary but not sufficient in general, so a complete higher-dimensional version of the $d=3$ shortcut remains open."],"forward_implications":["A matrix passing the symmetry-plus-eigenvalue test certifies that the identity rotation solves the associated Procrustes or Wahba alignment, without any SVD being run.","For $d=2$, the criterion is equivalent to the trace of the symmetric matrix being nonnegative, so the paper's closed-form rotation formula is optimal whenever the identity is not.","For $d=3$, maximality is equivalent to positive semidefiniteness of $\\operatorname{tr}(A)I-A$, a condition checkable by principal-minor tests before any eigenvector computation.","When a symmetric matrix fails the test, a rotation making it maximal is obtained by flipping the signs of its negative eigenvalues in an orthogonal diagonalization, with a parity rule deciding which eigenvalue's sign is preserved (Proposition 12).","For nonsymmetric $3\\times3$ matrices, the paper's Cayley-Newton procedure finds a rotation symmetrizing $UM$; the paper notes this procedure can fail when the required rotation has angle $\\pi$, leaving the SVD method as fallback."],"supporting_citations":[{"why":"present the SVD-based best-rotation algorithm that the paper's SVD-free methods are meant to complement.","marker":"[7, 8, 15]"},{"why":"supplies Proposition 3, the determinant-trace bound used in the sufficiency half of the characterization.","marker":"[9]"},{"why":"states the known orthogonal-matrix characterization (symmetric positive semidefinite) that this paper extends to rotations.","marker":"[6]"},{"why":"defines Wahba's problem, the weighted alignment problem whose solution is reduced to maximizing trace(UM).","marker":"[16]"},{"why":"supplies the singular value decomposition and eigenvector facts used in the proofs and in the 3D eigensolver.","marker":"[10]"},{"why":"provide the closed-form 3x3 symmetric eigensolver used for the SVD-free three-dimensional computation.","marker":"[5, 13]"}],"fun_headline_variants":["Maximal trace over rotations: symmetric with at most one negative eigenvalue","Eigenvalue sign rule decides if a matrix is rotation-optimal","Rotation-optimal matrices: symmetry plus a single signed eigenvalue condition","SVD-free check: symmetry and eigenvalue signs decide rotation-optimality"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof that the eigenvalue condition is sufficient leans on a technical bound: a reflection of a nonnegative diagonal matrix cannot raise the trace by more than twice the smallest diagonal entry, and the paper's printed derivation of the needed block structure contains a garbled determinant identity that a reader must repair. If that bound were false, the characterization would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Maximal trace over rotations: symmetric with at most one negative eigenvalue","Eigenvalue sign rule decides if a matrix is rotation-optimal","Rotation-optimal matrices: symmetry plus a single signed eigenvalue condition","SVD-free check: symmetry and eigenvalue signs decide rotation-optimality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000429,"raw_usage":{"total_tokens":2184,"prompt_tokens":928,"completion_tokens":1256,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":544,"completion_tokens_details":{"reasoning_tokens":1181}},"tokens_in":544,"tokens_out":1256,"duration_ms":8896,"temperature":1.0,"reasoning_tokens":1181,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:29:43.541158+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take $A=\\operatorname{diag}(1,-2)$. The paper's condition says $A$ should not be maximal over rotations; indeed, rotating by $\\pi$ gives $\\operatorname{tr}(UA)=1>-1=\\operatorname{tr}(A)$. To challenge the opposite direction, take $A=\\operatorname{diag}(1,-1/2)$ and check numerically over a fine grid of rotation angles that no rotation exceeds $\\operatorname{tr}(A)=1/2$. A decisive scan would randomly generate symmetric matrices, compute their eigenvalue conditions, and compare with a numerical maximization of $\\operatorname{tr}(UA)$ over $\\operatorname{SO}(d)$.","supporting_citations":[{"cited_title":"and Witzgall C., A Purely Algebraic Justification of the Kabsch-Umeyama Algorithm, Journal of Research of the National Institute of Standards and Technology 124 (2019) 1-6","cited_arxiv_id":null,"evidence_quote":"supplies Proposition 3, the determinant-trace bound used in the sufficiency half of the characterization."},{"cited_title":"and Johnson C., Matrix Analysis, Cambridge University Press, Cambridge 1985","cited_arxiv_id":null,"evidence_quote":"states the known orthogonal-matrix characterization (symmetric positive semidefinite) that this paper extends to rotations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"defines Wahba's problem, the weighted alignment problem whose solution is reduced to maximizing trace(UM)."},{"cited_title":"and McDonald J., Linear Algebra and Its Applications, 5th Edn, Pearson Education, Boston 2016","cited_arxiv_id":null,"evidence_quote":"supplies the singular value decomposition and eigenvector facts used in the proofs and in the 3D eigensolver."}],"review_version":1}