{"id":"96a299f0-441e-4f0a-9da1-e7b6f6b795ae","arxiv_id":"1908.03075","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"DDS 2.0 is a MATLAB solver for convex optimization over structured sets including quantum entropy, matrix norms, and hyperbolic polynomials, with certificates from an infeasible-start primal-dual method.","lead":"This paper describes DDS 2.0, a MATLAB package that solves a wide range of convex optimization problems using a single framework with many constraint types. It is useful for researchers working on quantum entropy, matrix norms, or hyperbolic programming, where existing modeling systems often rely on approximations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The QRE barrier in §9 is written with +ln(t−qre) rather than −ln, so as stated it is not a barrier at all; if the sign is a typo, the intended candidate is still explicitly 'not yet known to be s.c.', so QRE lacks the [19] convergence/certificate guarantee.","rationale":"The central claim is that DDS 2.0 solves every listed constraint class using the infeasible-start primal-dual algorithm from [19], with certificates. For that to hold, every constraint module must supply a self-concordant barrier and, for the primal-dual implementation, a computable conjugate or a justified primal-heavy substitute. The paper itself flags the QRE barrier as not yet known to be s.c.; the typeset formula makes the situation worse because the positive logarithm is not a barrier. I am not arguing that QRE problems cannot be solved numerically; the package and tables suggest they often are. The issue is that the central theorem-backed claim does not cover QRE as written, so the abstract overstates coverage. A sign correction and an explicit qualification of QRE as an empirical module would resolve the mismatch. This is the same broad weakness the reader identified, though the reader did not notice the sign problem; the conditional verdict is still appropriate after the check. I therefore do not change the reader's verdict.","tokens_in":42381,"tokens_out":8721,"duration_ms":104614,"concrete_test":"Inspect the QRE barrier routine in the released DDS source and evaluate it on the ray (t, I_2, I_2) as t↓0. A valid barrier increases to +∞; the paper's literal formula decreases to −∞. If the code matches the literal formula, the QRE module is not using a barrier; if it matches −ln(t−qre), the discrepancy is a documentation sign typo and only the unproven s.c. issue remains.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing weakness is in Section 9. The manuscript defines the quantum relative entropy barrier as Φ(t,X,Y) := ln(t − Tr(X ln X − X ln Y)) − ln det X − ln det Y, and immediately notes it is 'not yet known to be s.c.'. Two issues stack. (1) As typeset, the first term is +ln(t−qre(X,Y)), not −ln(t−qre(X,Y)). For the epigraph {qre(X,Y)≤t}, a barrier must tend to +∞ as t↓qre(X,Y); this expression tends to −∞, and +ln of a concave argument is not generally convex. Thus the displayed function is not a barrier for the stated set, independently of self-concordance. (2) If the missing minus sign is a typo and the code uses −ln(t−qre(X,Y))−ln det X−ln det Y, that candidate is precisely the one the authors flag as 'not yet known to be s.c.'. Either way the polynomial-time convergence, status certificates, and duality-gap analysis of [19,20] do not transfer to QRE constraints. The abstract's unconditional claim that DDS solves every listed constraint combination as an implementation of [19] is too strong; the QRE rows in Table 6 are empirical demonstrations, not guaranteed instances.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents DDS 2.0, a MATLAB package for convex optimization in the Domain-Driven form, i.e., problems of the form inf{<c,x> : Ax in D} where D is described by self-concordant barriers. The paper is primarily a user's guide, supplemented by theoretical appendices. It claims to accept every combination of eight constraint classes: symmetric cones, SOCP-representable quadratic constraints, 2D epigraphs of univariate convex functions, generalized power cones, matrix-norm epigraphs, vector relative entropy, quantum entropy and quantum relative entropy, and hyperbolic polynomials. The underlying algorithm is described as an infeasible-start primal-dual interior-point method from the authors' companion paper [19]. Numerical experiments compare DDS with CVX, CVXQUAD, and CBLIB/DIMACS benchmarks.","tokens_in":42657,"tokens_out":3820,"duration_ms":43509,"significance":"If the claims are correct, DDS 2.0 is a useful and unusually broad solver: it ships as working MATLAB code with many examples, it handles non-spectral convex constraints -- such as relative entropy, quantum entropy, and hyperbolic polynomial constraints -- without SDP approximations, and it provides detailed appendices with explicit formulas for gradients, Hessians, and Legendre-Fenchel conjugates. The use of external benchmark libraries (DIMACS, CBLIB) and comparisons against CVX and CVXQUAD for the quantum entropy cases are also strengths. However, the central claim of universal coverage of 'every combination' is compromised by the quantum relative entropy part: the displayed barrier is not a self-concordant barrier as written, and the paper explicitly states its self-concordance is not known. Consequently, the polynomial-time convergence and certificate guarantees from [19,20] do not transfer to QRE constraints, and the QRE numerical results are empirical demonstrations rather than guaranteed instances. This is a load-bearing weakness, not a presentation issue.","major_comments":[{"comment":"The displayed function Phi(t,X,Y) := ln(t - qre(X,Y)) - ln det X - ln det Y is not a barrier for the epigraph {qre(X,Y) <= t}. For a barrier one needs the value to tend to +infinity as t approaches qre(X,Y) from above; the first term, as written, tends to -infinity, and ln of a concave argument is not generally convex. If the intended function is -ln(t - qre(X,Y)) - ln det X - ln det Y, then the paper's immediately following statement that this function is 'not yet known to be s.c.' means that the main convergence and certificate theory of [19,20] does not apply to QRE constraints.","section":"Section 9, quantum relative entropy barrier"},{"comment":"The abstract and Section 1 state that DDS is a 'practical implementation' of the infeasible-start primal-dual algorithm of [19] and accepts every listed combination of constraints. For QRE, neither a self-concordance proof nor a computable Legendre-Fenchel conjugate is provided; Appendix C derives the conjugate only for quantum entropy, not for quantum relative entropy. Therefore the duality-gap stopping criteria (Section 13, Eqs. (71)-(74)) and the status certificates of [20] are not established for QRE rows in Table 6. The manuscript should explicitly qualify that QRE constraints are handled heuristically and that the corresponding numerical results are empirical, not covered by the theoretical guarantees.","section":"Section 9 and Table 6"}],"minor_comments":[{"comment":"In the paragraph after Eq. (24), 'GNC' appears where the context requires the abbreviation 'GPC' used elsewhere in the same section.","section":"Section 5, generalized power cone"},{"comment":"The text 'solving (DN) by [x,y]=DDS(c,A,b,Z)' should read '[x,y]=DDS(c,A,b,cons)' to match the calling syntax defined in Section 2.","section":"Section 6.1, nuclear norm"},{"comment":"The RHS for the SOCP block is given as 'b{2,2}' in one place, but the data structure consistently uses 'b{k,1}' for the shift vector; this should be corrected.","section":"Example 3.1, SOCP input"},{"comment":"Tables 5 and 6 report single runs with iteration counts and times but not final objective values, duality gaps, or accuracy relative to the stated tolerance; reporting these quantities would make the comparisons with CVXQUAD more informative and reproducible.","section":"Section 13, numerical results"}],"recommendation":"major_revision","confidential_remarks":"The QRE issue is substantive enough that the abstract's unconditional 'every combination' claim should not appear until the barrier is corrected and the theoretical status is clarified. The paper also leans heavily on the authors' own [19,20]; the appendices are largely original but the QRE gap must be addressed before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: DDS 2.0 is a serious MATLAB package with a broad constraint catalog, and the appendices contain real, useful derivations. The problem is Section 9. The quantum relative entropy barrier as displayed, Φ(t,X,Y)=ln(t−Tr(XlnX−XlnY))−ln det X−ln det Y, cannot be a barrier for {qre≤t}: the first term tends to −∞ as t approaches the boundary, and +ln of a concave argument is not convex. If the intended minus sign is a typo, the corrected candidate is precisely the one the authors mark as 'not yet known to be s.c.', so the convergence and certificate guarantees from [19] and [20] do not apply to QRE constraints. Either way, the abstract's claim that DDS solves every listed constraint combination as an implementation of [19] overstates the case; the QRE rows in Table 6 are empirical.\n\nNow the credit. The package genuinely expands the domain-driven framework: symmetric cones, SOCP-representable quadratics, generalized power cones, matrix norm epigraphs, vector relative entropy, quantum entropy, hyperbolic polynomials. The numerical comparisons against CVX and CVXQUAD on DIMACS and CBLIB problems are reasonable, and they report infeasible cases, which is more than most package papers do. The proof in Appendix E that the vector relative entropy barrier is (2ℓ+1)-self-concordant is a standalone result I'd want to have on record. Appendix D's explicit Legendre-Fenchel conjugates and derivative formulas for the univariate barriers are exactly what an implementer needs; that is real work. The code is public, which is reproducible in principle.\n\nSoft spots besides the QRE issue: the experiments are single runs with no accuracy table, and there is no commit hash for the code, so I can't verify which QRE barrier the code actually uses. Those are minor by package-paper standards. The load-bearing issue is the QRE guarantee, and it needs to be fixed in the text and likely in the abstract. The authors are transparent about the missing self-concordance, which I respect, but transparency doesn't turn an empirical result into a certified one.\n\nWho should read this: anyone implementing interior-point methods for exotic cones, or trying to solve entropy/quantum-entropy problems without SDP approximations. The vector relative entropy proof alone makes the paper worth a place in the literature.\n\nMy recommendation: send it to peer review with a request to fix the QRE sign, state clearly which QRE claims are empirical, and adjust the abstract's universal promise. The rest of the paper is solid enough to justify the referee time.","headline":"A genuinely useful solver package with real new derivations, but the quantum relative entropy section has a sign error that makes the displayed barrier no barrier at all, and even the intended candidate lacks the self-concordance guarantee the rest of the theory relies on.","tokens_in":43174,"tokens_out":3133,"would_cite":false,"duration_ms":34605,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["90C25","90C51","90C22"],"pacs":[],"model":"deepseek-v4-flash","headline":"DDS 2.0 solves every combination of eight convex constraint families with one infeasible-start primal-dual interior-point method and returns certificates for the main statuses.","keywords":["convex optimization","domain-driven form","self-concordant barriers","interior-point methods","quantum relative entropy","generalized power cone","hyperbolic programming","MATLAB solver"],"falsifier":"Test the required smoothness inequality for the quantum relative entropy barrier -ln(t - Tr(X ln X - X ln Y)) - ln det X - ln det Y at nearly commuting positive-definite matrices X and Y: if some feasible direction h makes the third derivative exceed twice the 3/2 power of the second derivative, the barrier is not of the needed type and the polynomial-time guarantee for QRE collapses.","tokens_in":42148,"feed_emoji":"🧮","tokens_out":11901,"duration_ms":119882,"temperature":0.7,"pith_summary":"The paper presents DDS 2.0, a MATLAB package for convex optimization problems written in Domain-Driven form: minimize a linear function of x subject to Ax lying in a convex set D. The version described here accepts every combination of symmetric cones (LP, SOCP, SDP), SOCP-representable quadratic constraints, epigraphs of univariate convex functions, generalized power cones, matrix-norm epigraphs, vector relative entropy, quantum entropy, quantum relative entropy, and hyperbolic-polynomial constraints. The solver implements an infeasible-start primal-dual interior-point method and pairs each constraint family with a self-concordant barrier, plus its Legendre-Fenchel conjugate whenever that conjugate can be evaluated efficiently. A sympathetic reader should care because this is one code path that returns certificates distinguishing approximately optimal, infeasible, unbounded, and ill-conditioned cases for a much wider class of non-conic convex problems than standard conic solvers handle directly. The manuscript is primarily a user's guide with implementation details and numerical experiments on the newer constraint families.","feed_headline":"Eight convex constraint families meet in one solver, DDS 2.0","feed_subtitle":"Each constraint gets a tailored self-concordant barrier, so dual certificates stay exact rather than approximated.","key_machinery":"The load-bearing object is the Domain-Driven form, in which the feasible set is written as Ax in D with D the closure of the domain of a self-concordant barrier, a smooth penalty that blows up at the boundary and whose curvature changes slowly enough for interior-point methods to be provably efficient. The algorithm alternates predictor and corrector steps built from the barrier's gradient and Hessian together with the Legendre-Fenchel conjugate of the barrier, which is the companion dual penalty used to form search directions and to evaluate the support function needed for certificates. Specific barriers carry the individual constraint families: matrix-monotone functions give the quantum-entropy barrier, a proven barrier handles the generalized power cone, a theorem for hyperbolic polynomials gives -ln(p(x)) on the hyperbolicity cone, and the paper proves a self-concordant barrier for vector relative entropy in an appendix. For quantum relative entropy, DDS uses the barrier -ln(t - Tr(X ln X - X ln Y)) - ln det X - ln det Y, which the paper explicitly marks as not yet known to be self-concordant, so the theoretical convergence guarantees do not yet cover that family.","core_discovery":"The paper's thesis is that the Domain-Driven formulation is not only a theoretical umbrella but a workable software design: one solver can accept every combination of the listed constraint families and solve them with a single infeasible-start primal-dual interior-point method, provided each family supplies a self-concordant barrier and, where possible, its Legendre-Fenchel conjugate. For each family, DDS uses the barrier tailored to that set rather than reformulating the set as a semidefinite program, which lets it return dual certificates for problems whose constraints are not spectrahedral. The stopping rules and status reports are taken from a companion analysis of status determination, so the solver distinguishes solved, infeasible, unbounded, and ill-conditioned outcomes at a user-chosen tolerance. The numerical sections show the direct-barrier approach scaling well on nuclear-norm, quantum-entropy, and hyperbolic-polynomial problems, including a primal-heavy variant used when the conjugate barrier is unavailable.","pith_inferences":["Editorial inference: if a self-concordant barrier for quantum relative entropy is later found, the existing code could absorb it without structural change, and the numerical QRE results would move from empirical demonstrations to certified solutions.","Editorial inference: the certificate machinery suggests a testable use outside the package: on a suspected-infeasible relative-entropy instance, one can independently check the returned dual certificate, giving a practical oracle for convex infeasibility in information-theoretic relaxations.","Editorial inference: the same barrier-plus-conjugate recipe points to an expansion path for future constraint families; whenever a new barrier has a conjugate that reduces to a few one-dimensional root solves, that family can likely be added to DDS with modest code changes.","Editorial inference: if the QRE barrier is not self-concordant, the QRE entries in the numerical tables should be interpreted as empirical performance data, and the status certificates for those runs would need independent verification before being relied on."],"forward_implications":["Because all listed constraint families can be mixed in one problem, a user can model a convex program with entropy, quantum, and conic constraints together and solve it without converting the whole problem to a standard symmetric cone.","For families with proven barriers, the solver inherits the polynomial iteration bound of the infeasible-start primal-dual algorithm, so the practical stopping rules are backed by a convergence guarantee rather than only by heuristics.","Direct use of tailored barriers avoids semidefinite approximations for non-spectral constraints; the numerical experiments indicate near-linear growth in running time for nuclear-norm minimization when the matrix is much wider than tall.","Hyperbolic-polynomial constraints, which standard modeling systems generally do not accept, become solvable through the primal-heavy variant that approximates the missing conjugate while keeping dual iterates feasible."],"supporting_citations":[{"why":"Supplies the infeasible-start primal-dual algorithm that DDS implements, including the iteration-complexity analysis and duality-gap framework.","marker":"[19]"},{"why":"Provides the status-determination and stopping-criteria theory used by DDS to report solved, infeasible, unbounded, or ill-conditioned.","marker":"[20]"},{"why":"Gives the self-concordant barrier theory and compatibility results used to construct and prove barriers for the constraint families.","marker":"[27]"},{"why":"Shows how matrix-monotone functions give self-concordant barriers for quantum entropy epigraphs.","marker":"[14]"},{"why":"Proves that the generalized power cone barrier used in DDS is self-concordant.","marker":"[31]"},{"why":"Establishes that -ln(p(x)) is a self-concordant barrier on the hyperbolicity cone of a hyperbolic polynomial.","marker":"[17]"},{"why":"Provides barrier and conjugate formulas for matrix-norm epigraphs and the cone-free primal-dual methodology underlying the algorithm.","marker":"[25]"}],"fun_headline_variants":["One solver, eight constraint families, no SDP reformulation","DDS 2.0: exact duals for non-spectrahedral convex sets","Direct barriers beat SDP reformulation in DDS 2.0","Convex solver handles quantum entropy and hyperbolic polynomials","DDS 2.0: infeasible-start primal-dual for all domain-driven forms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The convergence and certificate guarantees rest on every constraint family having a smooth penalty function of the special type interior-point methods require, and for quantum relative entropy the paper explicitly says that property is not yet known to hold.","fun_headline_variants_meta":{"raw":{"variants":["One solver, eight constraint families, no SDP reformulation","DDS 2.0: exact duals for non-spectrahedral convex sets","Direct barriers beat SDP reformulation in DDS 2.0","Convex solver handles quantum entropy and hyperbolic polynomials","DDS 2.0: infeasible-start primal-dual for all domain-driven forms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000141,"raw_usage":{"total_tokens":1194,"prompt_tokens":1005,"completion_tokens":189,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":92}},"tokens_in":621,"tokens_out":189,"duration_ms":2949,"temperature":1.0,"reasoning_tokens":92,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:36:07.847888+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Test the required smoothness inequality for the quantum relative entropy barrier -ln(t - Tr(X ln X - X ln Y)) - ln det X - ln det Y at nearly commuting positive-definite matrices X and Y: if some feasible direction h makes the third derivative exceed twice the 3/2 power of the second derivative, the barrier is not of the needed type and the polynomial-time guarantee for QRE collapses.","supporting_citations":[{"cited_title":"Primal-Dual Interior-Point Methods for Domain-Driven Formulations","cited_arxiv_id":"1804.06925","evidence_quote":"Supplies the infeasible-start primal-dual algorithm that DDS implements, including the iteration-complexity analysis and duality-gap framework."},{"cited_title":"Status Determination by Interior-Point Methods for Convex Optimization Problems in Domain-Driven Form","cited_arxiv_id":"1901.07084","evidence_quote":"Provides the status-determination and stopping-criteria theory used by DDS to report solved, infeasible, unbounded, or ill-conditioned."},{"cited_title":"Nesterov and A","cited_arxiv_id":null,"evidence_quote":"Gives the self-concordant barrier theory and compatibility results used to construct and prove barriers for the constraint families."},{"cited_title":"F aybusovich and T","cited_arxiv_id":null,"evidence_quote":"Shows how matrix-monotone functions give self-concordant barriers for quantum entropy epigraphs."},{"cited_title":"Roy and L","cited_arxiv_id":null,"evidence_quote":"Proves that the generalized power cone barrier used in DDS is self-concordant."},{"cited_title":"G ¨uler, Hyperbolic polynomials and interior point methods for conv ex programming, Mathematics of Op- erations Research, 22 (1997), pp","cited_arxiv_id":null,"evidence_quote":"Establishes that -ln(p(x)) is a self-concordant barrier on the hyperbolicity cone of a hyperbolic polynomial."},{"cited_title":"Nemirovski and L","cited_arxiv_id":null,"evidence_quote":"Provides barrier and conjugate formulas for matrix-norm epigraphs and the cone-free primal-dual methodology underlying the algorithm."}],"review_version":1}