{"id":"c420d464-9405-4e31-803f-71953f4e4ddd","arxiv_id":"1908.09211","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper derives inequalities and conditional equalities linking Kantorovich-Wasserstein transport cost to Kullback-Leibler divergence through the optimal channel problem and a Pythagorean identity.","lead":"This paper connects two ways of comparing probability distributions: the optimal transport cost and the Kullback-Leibler divergence. It shows transport is a constrained version of a rate-distortion problem, so the rate-distortion cost bounds the transport cost from below.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The exact KL–KW equality in Theorem 2 requires a common reference measure r for which the optimal dual potentials are log-density ratios; this condition is uncharacterized and, for metric costs, forces ln(dp/dq) ≤ 0 pointwise, so the advertised exact link generically reduces to an inequality.","rationale":"The paper's main positive contribution is the observation that fixing both marginals makes OTP a constrained version of the optimal-channel problem, so Rc[q](λ) ≤ Kc[p,q](λ). This is correct and is properly credited in the reader's verdict. The algebraic identities (7)-(8) are also correct. The risk is concentrated where the paper promotes inequalities to equalities. Theorem 2's exact decomposition depends on a common-reference gradient representation that is not automatic; the diagonal feasibility condition shows it fails unless a pointwise dominance condition holds. Since the paper both claims 'the link between two divergences' and acknowledges inequalities generally, the exact equality is the load-bearing part of the advertised result. The reader identified the same weakness, and the proposed LP test on a two-point example would make the restriction concrete. The Theorem 1 proof's subdifferential assertion is a further reason not to accept the equality claims as written. None of this undermines the lower-bound result, so the appropriate verdict remains conditional.","tokens_in":6978,"tokens_out":10849,"duration_ms":104745,"concrete_test":"On X=Y={0,1}, take c(0,0)=c(1,1)=0, c(0,1)=c(1,0)=1, p=(0.8,0.2), q=(0.2,0.8). Solve the dual LP (6) to obtain optimal potentials up to gauge and check whether any gauge allows f = ln(dp/dr) and g = ln(dq/dr) for a common r. A necessary condition is ln(dp/dq)(x) ≤ c(x,x)=0 for x=0,1, i.e. ln 4 ≤ 0, which fails. Hence no such r exists, and Theorem 2's exact equality does not apply to this generic pair; the paper's formula then yields only the inequality with ε>0. This settles whether the common-reference assumption is automatic or restrictive.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central exact claim is Theorem 2: if the Kantorovich dual optimizers (f,g) satisfy f = ln(dp/dr) and g = ln(dq/dr) for a common r, then D[p,q] = Kc[p,q] - (κ[f]-κ[g]) - ∫ g(dp-dq). The condition is not a harmless normalization. Because f and g are log-density ratios, f(x)-g(x) = ln(dp/dq)(x). Feasibility in (6) requires f(x)-g(y) ≤ c(x,y) for all x,y; with y=x this forces ln(dp/dq)(x) ≤ c(x,x). For metric costs c(x,x)=0 the equality regime requires p/q ≤ 1 a.e., so pairs such as p=(0.8,0.2), q=(0.2,0.8) with a 0/1 metric cost are excluded. The paper gives no construction of r and no characterization of when the condition holds, so the equality in Theorem 2 is not a general link. The companion equality claim in Theorem 1 is also not established: its proof asserts that any equal-cost wOTP lies in the subdifferential ∂D*[-βc,q⊗p], but the optimal channel (4) is not the unconstrained tilted product q⊗p e^{-βc}, and ∂D* is a singleton, so the argument does not justify the claimed iff. What remains solid is the lower bound Rc[q](λ) ≤ Kc[p,q](λ), which follows directly from constraint relaxation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the relation between the Kantorovich-Wasserstein cost Kc[p,q] defined in (1) and the KL divergence D[p,q] defined in (2). It introduces the optimal channel problem Rc[q](λ) in (3), which fixes only the input marginal and constrains mutual information, and observes that Rc[q](λ) ≤ Kc[p,q](λ) because the transport problem additionally fixes the output marginal. Theorem 1 claims that equality holds if and only if the two problems share the same joint optimizer, with a proof based on subdifferentials of the log-partition functional D*[u,q⊗p]. The paper also derives the Pythagorean identity (5) and the law-of-cosines identities (7)-(8), and uses the dual transport problem (6) to prove Theorem 2: if the optimal dual potentials f,g are log-density ratios against a common reference measure r, then D[p,q] = Kc[p,q] − (κ[f] − κ[g]) − ∫ g(dp−dq). The Discussion states that generally only inequalities are obtained.","tokens_in":7265,"tokens_out":13394,"duration_ms":144336,"significance":"The lower-bound inequality Rc[q](λ) ≤ Kc[p,q](λ) is correct and follows immediately from constraint relaxation, and the algebraic identities (5), (7), and (8) are correct by substitution. The paper is honest in its Discussion that the general relations are inequalities. However, the exact equality results are the main advertised contribution, and they are either false as stated (Theorem 1) or conditional on an uncharacterized and very restrictive hypothesis (Theorem 2). The solid remainder of the paper is an elementary relaxation inequality plus a conditional algebraic identity, so the significance is substantially lower than the abstract and theorems suggest.","major_comments":[{"comment":"The claimed characterization of equality is false, and the proof is not valid. The assertion that wOCP belongs to ∂D*[−βc,q⊗p] does not follow from the definition of the OCP in (3): the OCP imposes only π_X w=q and I(X;Y)≤λ, and it is not the unconstrained tilted-problem whose solution is the normalized q⊗p e^{−βc}; note that the normalization κ(β,x) in (4) is x-dependent. Moreover, ∂D* is a singleton when it exists, so the convex-combination argument is irrelevant. A concrete counterexample to the 'only if' direction is c≡0 on a finite product space: for any q≠p with I(q⊗p)≤λ one has Rc[q](λ)=Kc[p,q](λ)=0, but wOCP=q⊗q and wOTP=q⊗p are distinct optimal solutions. The valid and elementary statement is the inequality Rc[q](λ)≤Kc[p,q](λ), obtained by constraint relaxation.","section":"Section 2, Theorem 1"},{"comment":"The equality in Theorem 2 is proved correctly by substitution, but the hypothesis is not characterized or even discussed. No construction or existence condition for the reference measure r is given, and for typical costs the hypothesis excludes most pairs (p,q). For a metric cost with c(x,x)=0, feasibility f(x)−g(y)≤c(x,y) together with f=ln(dp/dr) and g=ln(dq/dr) gives ln(dp/dq)(x)≤0 pointwise at y=x, i.e., p≤q. Thus the advertised exact link is not generic; without a characterization of when the hypothesis is satisfiable, Theorem 2 remains a conditional algebraic identity rather than a relation between the two divergences. The paper should provide such a characterization with concrete examples, or explicitly present Theorem 2 as an identity holding only under a strong extra condition.","section":"Section 3, Theorem 2"},{"comment":"Even apart from the subdifferential membership, the proof conflates equality of optimal values with equality of optimizers. The fact that an optimal channel achieves the same cost as the optimal transport plan does not imply that the optimal channel is feasible for the transport problem or that the transport plan is optimal for the channel problem; equality of infima can occur with different minimizers. This is the root of the c≡0 counterexample above. The theorem needs additional uniqueness or feasibility assumptions before any 'if and only if' statement can be made.","section":"Section 2, proof of Theorem 1"}],"minor_comments":[{"comment":"The identity D[w,q⊗q]=D[w,q⊗p]+D[p,q] is valid only when p is the output marginal π_Y w, or more generally when ∫ ln(dp/dq) dw = ∫ ln(dp/dq) dp. Please state this assumption explicitly, because as written the identity appears to hold for arbitrary w.","section":"Section 2, equation (5)"},{"comment":"The notation κ[f] is used in Theorem 2 although κ was previously defined for βf as κ[βf]. Since the theorem assumes β=1, either define κ directly or note that when f=ln(dp/dr), one has κ[f]=0; the displayed formula is then unnecessarily opaque.","section":"Section 3, Theorem 2"},{"comment":"The notation ∇D[p,r] is informal: the KL divergence is a functional of measures, so the derivative is a functional derivative. Please clarify that ∇D[p,r] means the log-density ln(dp/dr).","section":"Section 3, display before Theorem 2"},{"comment":"The symbol p is used both as a fixed measure in the product q⊗p and as the output marginal of w in the same displayed formula. This is confusing; use distinct symbols for the fixed output measure and the induced output marginal.","section":"Section 2, equation (4)"},{"comment":"The proof of Theorem 1 relies on a standard Lagrange-multiplier result 'e.g. see [12, 2]', but no specific theorem is quoted. Since the subdifferential claim is central, please state the relevant result and verify that its hypotheses are satisfied by the OCP in (3).","section":"References"}],"recommendation":"major_revision","confidential_remarks":"This is a short note whose central inequality is correct but elementary; the main equality claims are not. Theorem 1 is false as stated and Theorem 2 is conditional on an uncharacterized and restrictive hypothesis. I recommend major revision rather than rejection because the paper contains correct algebraic identities and the honest inequality; however, the authors must either repair Theorem 1 with explicit uniqueness/feasibility assumptions or remove it, and they must characterize or carefully reframe Theorem 2. If that cannot be done, the paper will likely fall below the standard for this journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the core observation is correct but elementary, and the attempt to promote it to an exact equivalence runs into real problems. As a note, it deserves a referee; as a theorem paper, it overreaches.\n\nThe new bit that works is taking the Kantorovich OTP and noticing that fixing the output marginal is exactly the extra constraint that separates it from the rate-distortion / optimal channel problem. The inequality Rc[q](λ) ≤ Kc[p,q](λ) follows immediately by relaxing that constraint, and the paper states it plainly. The law-of-cosines decomposition in Section 3 is algebraically correct, and the discussion is appropriately cautious: it says the relations generally hold as inequalities.\n\nThe soft spots are in the equality claims. Theorem 1's proof asserts that the optimal channel lies in the subdifferential of the log-partition functional at −βc. But the OCP's information constraint involves the output marginal of w itself, not a fixed product measure, so the tilted product used in the proof isn't the OCP solution in general. The subdifferential is also a singleton, so the argument that any equal-cost transport plan would also lie in it is not available. Theorem 1's statement may be true for a different reason—if both problems share a solution, the values obviously agree—but the proof as written doesn't establish the 'only if'.\n\nTheorem 2 is a bigger issue. The condition that the optimal dual potentials are log-likelihood ratios against one common reference r, f=ln(dp/dr), g=ln(dq/dr), is not harmless. Since the dual feasibility requires f(x)−g(y) ≤ c(x,y), taking y=x gives ln(dp/dq)(x) ≤ c(x,x). For the metric costs the KW metric is built from, c(x,x)=0, so the condition forces p≤q pointwise, hence p=q. The equality regime is essentially empty in the intended application. The paper gives no construction of r or characterization of when the condition holds, so Theorem 2 reduces to an inequality with an extra assumption that is not understood.\n\nWhat remains is a correct lower bound and a nice geometric picture. The paper doesn't overclaim in the abstract, and the discussion is honest. A referee could repair Theorem 1 by proving the iff directly from feasibility, and could downgrade Theorem 2 to a conditional equality with a worked example. As a short note in a workshop or conference proceedings this might be fine; as a journal paper it needs revision. I'd send it to review, but with the expectation of major changes.","headline":"Correct but elementary lower-bound observation; the equality theorems are overclaimed and Theorem 2's condition is nearly vacuous for metric costs.","tokens_in":7819,"tokens_out":5064,"would_cite":false,"duration_ms":53814,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["49Q22","94A17"],"pacs":[],"model":"deepseek-v4-flash","headline":"The optimal-transport metric is the optimal channel cost with the output distribution fixed; under a gradient condition, the KL divergence decomposes into it plus corrections.","keywords":["optimal transport","Wasserstein metric","Kantorovich metric","Kullback-Leibler divergence","rate-distortion theory","value of information","optimal channel","law of cosines"],"falsifier":"For a three-point set with a generic cost matrix and two probability vectors $p,q$, solve the Kantorovich dual (6) to get potentials $f,g$, then check whether any reference measure $r$ satisfies $p_i = e^{f_i}r_i / Z_p$ and $q_i = e^{g_i}r_i / Z_q$. If no such $r$ exists, the exact identity cannot hold for that instance, and only the inequality remains.","tokens_in":6730,"feed_emoji":"🚚","tokens_out":12104,"duration_ms":114792,"temperature":0.7,"pith_summary":"This paper argues that the optimal-transport distance between probability measures and the Kullback-Leibler divergence are connected rather than unrelated. Its main claim is that the optimal transport problem is the same as the rate-distortion optimal channel problem except for one extra constraint: the output distribution is fixed in advance. As a consequence, the optimal channel cost is always a lower bound on the optimal-transport cost, with equality exactly when the two problems share the same optimal joint distribution. The paper then uses a geometric law-of-cosines identity for the KL divergence to decompose $D[p,q]$ into the transport cost plus terms involving the dual potentials, under the condition that those potentials are log-likelihood ratios with respect to a common reference measure. If the argument holds, it gives a bridge between optimal-transport geometry and information geometry, with rate-distortion computations becoming usable as bounds on transport costs.","feed_headline":"Optimal transport is rate distortion with a fixed output","feed_subtitle":"The optimal channel cost is a lower bound on the Wasserstein metric; equality needs a log-likelihood gradient condition.","key_machinery":"The argument is carried by two devices. The first is the observation that the two-marginal optimal transport problem is exactly an optimal channel problem with the output marginal fixed: the rate-distortion channel has the Gibbs form $dw(x,y)=dq(x)\\,dp(y)e^{-\\beta c(x,y)-\\kappa(\\beta,x)}$, and its value $R_c[q](\\lambda)$ competes with the transport value $K_c[p,q](\\lambda)$. The second is the law of cosines for the KL divergence, $D[p,q]=D[p,r]+D[r,q]-\\int\\ln(dq/dr)(dp-dr)$, which, using the exponential representations $dp=e^{f-\\kappa[f]}dr$ and $dq=e^{g-\\kappa[g]}dr$, turns the dual transport objective $E_p\\{f\\}-E_q\\{g\\}$ into a component of the KL divergence.","core_discovery":"The central claim is that Kantorovich's optimal transport problem is the optimal channel (rate-distortion) problem with one additional constraint fixing the output marginal. Written with the same information constraint $I(X,Y)\\leq\\lambda$, the optimal channel value $R_c[q](\\lambda)$ is always a lower bound on the transport value $K_c[p,q](\\lambda)$, and the two are equal exactly when the same joint measure solves both problems. On the geometric side, the paper uses the law of cosines for the KL divergence to decompose $D[p,q]$ as a difference of expectations plus a correction term. If the optimal dual functions $f,g$ from the Kantorovich dual are themselves KL gradients with respect to a common reference measure $r$, then the decomposition becomes the exact additive identity $D[p,q]=K_c[p,q]-(\\kappa[f]-\\kappa[g])-\\int g\\,(dp-dq)$. Without that condition, only an inequality $D[p,q]\\leq \\epsilon K_c[p,q]-(\\kappa[\\beta f]-\\kappa[\\alpha g])-\\alpha\\int g\\,(dp-dq)$ is available.","pith_inferences":["The equality condition is likely to fail for generic non-exponential-family measures, so the practical payload of the paper is the inequality: rate-distortion lower bounds on Wasserstein distances that remain computable when the transport problem is hard.","The common-reference requirement suggests that exact equality holds precisely along exponential families, making the identity an information-geometric characterization of transport duality.","A numerical study on Gaussian mixtures or discrete graphical models could quantify the gap between $R_c[q](\\lambda)$ and $K_c[p,q](\\lambda)$ and reveal how tight the bound is.","The game-theoretic point that optimal channels beat transport plans when the output is unconstrained points toward stochastic maps that minimize expected cost without a fixed target output."],"forward_implications":["For every output measure $p$, the rate-distortion value $R_c[q](\\lambda)$ is a lower bound on the optimal-transport cost $K_c[p,q](\\lambda)$, so optimal channel computations give certificates for transport costs.","When the equality conditions hold, the optimal-transport cost can be recovered from a Kullback-Leibler-constrained optimization, connecting rate-distortion algorithms to optimal-transport algorithms.","The identity $D[w,q\\otimes q]=I(X,Y)+D[p,q]$ converts the information constraint into a constraint on the divergence between the marginals, giving an alternative description of the constrained transport value.","For translation-invariant costs, the optimal channel simplifies to a form independent of the input marginal, so the corresponding transport cost depends on the output measure explicitly."],"supporting_citations":[{"why":"defines the Kantorovich optimal transport problem whose value $K_c$ is the paper's central object.","marker":"[7]"},{"why":"introduces the KL divergence $D[p,q]$ that the paper compares with the transport cost.","marker":"[8]"},{"why":"formulates the rate-distortion optimal channel problem $R_c[q](\\lambda)$.","marker":"[10]"},{"why":"extends the channel problem to value of information, giving the optimal channel context.","marker":"[11]"},{"why":"supplies the law-of-cosines decomposition of KL divergence used in equation (7).","marker":"[1]"},{"why":"supports the representation of the optimal channel as a subdifferential of the convex dual functional.","marker":"[2]"},{"why":"provides the Lagrange-multiplier method behind the exponential form of the optimal channel.","marker":"[12]"}],"fun_headline_variants":["Fixed output turns rate distortion into optimal transport","Rate-distortion cost is a lower bound on Wasserstein distance","Optimal transport and rate distortion differ by one constraint","KL cosine law unifies Wasserstein metric and rate distortion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The exact equality rests on the assumption that the two optimal dual functions are log-likelihood ratios of the two measures against the same reference measure; for arbitrary measures and costs, no such reference measure is guaranteed to exist.","fun_headline_variants_meta":{"raw":{"variants":["Fixed output turns rate distortion into optimal transport","Rate-distortion cost is a lower bound on Wasserstein distance","Optimal transport and rate distortion differ by one constraint","KL cosine law unifies Wasserstein metric and rate distortion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00034,"raw_usage":{"total_tokens":1866,"prompt_tokens":929,"completion_tokens":937,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":545,"completion_tokens_details":{"reasoning_tokens":873}},"tokens_in":545,"tokens_out":937,"duration_ms":8424,"temperature":1.0,"reasoning_tokens":873,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:19:21.628671+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a three-point set with a generic cost matrix and two probability vectors $p,q$, solve the Kantorovich dual (6) to get potentials $f,g$, then check whether any reference measure $r$ satisfies $p_i = e^{f_i}r_i / Z_p$ and $q_i = e^{g_i}r_i / Z_q$. If no such $r$ exists, the exact identity cannot hold for that instance, and only the inequality remains.","supporting_citations":[{"cited_title":"USSR AS Do klady 37(7–8), 227–229 (1942)","cited_arxiv_id":null,"evidence_quote":"defines the Kantorovich optimal transport problem whose value $K_c$ is the paper's central object."},{"cited_title":"The Annals of Math- ematical Statistics 22(1), 79–86 (1951)","cited_arxiv_id":null,"evidence_quote":"introduces the KL divergence $D[p,q]$ that the paper compares with the transport cost."},{"cited_title":"Bell System Technical Journal 27, 379–423 and 623–656 (1948)","cited_arxiv_id":null,"evidence_quote":"formulates the rate-distortion optimal channel problem $R_c[q](\\lambda)$."},{"cited_title":"Izvestiya of USSR Academy of Sciences, Technical Cybernetics 5, 3–12 (1965)","cited_arxiv_id":null,"evidence_quote":"extends the channel problem to value of information, giving the optimal channel context."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the law-of-cosines decomposition of KL divergence used in equation (7)."},{"cited_title":"Journal of Global Optimization 55, 387–416 (2013)","cited_arxiv_id":null,"evidence_quote":"supports the representation of the optimal channel as a subdifferential of the convex dual functional."},{"cited_title":"Sovetskoe R adio, Moscow, USSR (1975)","cited_arxiv_id":null,"evidence_quote":"provides the Lagrange-multiplier method behind the exponential form of the optimal channel."}],"review_version":1}