{"id":"2b1bff7b-ef50-4e2f-86d3-0bc86c4b89ad","arxiv_id":"2505.13351","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A canonical linear Poisson bracket is constructed on the predual bundle of a Banach Lie algebroid, with necessary and sufficient conditions for it to define a Banach Poisson manifold.","lead":"Mathematicians show that the predual bundle of a Banach Lie algebroid, a curved infinite-dimensional space, carries a natural linear Poisson bracket. This is a cleaner infinite-dimensional analogue of a classical finite-dimensional construction and clarifies exactly when Hamiltonian dynamics is well defined.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sign error in Lemma 3.3: the sharp map formula contradicts the defining relations (3.4); the proof reverses the pairing order, so the proof of Theorem 3.6 is invalid as written.","rationale":"After reading the construction, the main claim is plausible: the local bracket (3.5) is a genuine linear Poisson bracket candidate, and the two inclusion conditions are the natural nonsingularity requirements. The reader's concern about the non-queer assumption is a scoping caveat, not a defect of the argument within its stated assumptions. The Jacobi identity is asserted rather than proved, but the reduction to the spanning family can be made rigorous via the Schouten tensor if the predual setting allows it, so this is a gap in exposition rather than a demonstrated error. The sign error in Lemma 3.3 is a concrete inconsistency: the sharp map displayed there does not satisfy the relations (3.4) that define the Poisson tensor. Since Lemma 3.3 is the basis for Theorem 3.6, the proof of the central 'if and only if' is not trustworthy as written. The error is easily corrected and does not change the inclusion conditions, so the appropriate verdict remains CONDITIONAL pending this correction.","tokens_in":13250,"tokens_out":38376,"duration_ms":369549,"concrete_test":"Recompute the sharp map by inserting the local bracket (3.5) into the definition ♯(α)(β)=Π(α,β): set α=dλ_X and β=d(f∘π_*) and check whether the resulting vector is (a(x), −a^*(µ)+(ad^x)^*φ) or its negative. If the corrected formula holds, Theorem 3.6's conditions are unchanged; if the manuscript's formula is used, the sign of {λ_X,f∘π_*} becomes −(a(X)f), contradicting (3.4).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Lemma 3.3 claims ♯(m,φ,µ,x)=(m,φ,−a(x), a^*(µ)−(ad^m_x)^*φ). Consistency with the defining bracket (3.4) requires ♯(dλ_X)(d(f∘π_*))={λ_X,f∘π_*}=(a(X)f)∘π_*. In the local trivialization, dλ_X=(d(φ∘v), v) and d(f∘π_*)=(df,0), so the claimed ♯ gives ⟨−a(v), df⟩=−(a(X)f), the opposite sign. The proof of Lemma 3.3 reverses the order of the sharp map and its argument: it writes ♯(df,0)(dλ_v)={λ_v,f∘π_*}, whereas ♯(α)(β)={f,g} with α=df, β=dg yields ♯(df,0)(dλ_v)={f∘π_*,λ_v}. The correct sharp map reproducing the local bracket (3.5) is ♯(µ,x)=(a(x), −a^*(µ)+(ad^m_x)^*φ). Because E_* is a linear subspace, the sign change leaves the membership conditions (3.6)–(3.7) invariant, so Theorem 3.6's statement may survive, but its proof is not valid as written.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a linear Poisson bracket on the predual bundle E_* of a Banach Lie algebroid E over a Banach manifold M, assuming the existence of a smooth predual bundle and excluding 'queer' algebroids. Section 3 defines the bracket by the relations (3.4), computes the sharp map in Lemma 3.3, gives the local coordinate formula (3.5), and states necessary and sufficient conditions for E_* to be a Banach Poisson manifold in the sense of Definition 2.1 (Theorem 3.6). Section 4 proposes a converse construction of a Lie algebroid on E from a linear Poisson bracket on E_*. Sections 5 and 6 discuss the precotangent bundle example and a trivial ℓ2×ℓ∞ example. The construction is explicit and parameter-free, and it reduces to the known Banach Lie–Poisson space theorem [OR03] when M is a point.","tokens_in":13493,"tokens_out":42550,"duration_ms":394200,"significance":"If the main construction is correct, the paper gives a clean infinite-dimensional analogue of the linear Poisson structure on the dual of a Lie algebroid, with a concrete criterion for the existence of Hamiltonian vector fields. The explicit local formula (3.5) and the reduction to the Lie algebra case are useful sanity checks. The paper also contributes a partial negative result on queer Banach Lie algebroids for fibers with a Schauder basis. However, the manuscript as written has a sign error in the converse theorem of Section 4 and an insufficiently justified Jacobi-identity step in Theorem 3.2, so the claims are not yet fully supported.","major_comments":[{"comment":"The anchor as defined in Theorem 4.3 has a sign inconsistency. With Definition 2.1, ♯(dλ_X)(d(f∘π_*)) = {f∘π_*, λ_X}; the paper itself uses this convention in the proof of Lemma 3.3. Therefore for a(X) = Tπ_*♯(dλ_X), one has (a(X)f)∘π_* = {f∘π_*, λ_X} = -{λ_X, f∘π_*}. In the computation of [X_1, fX_2], the term {λ_{X_1}, f∘π_*} is replaced by ♯(dλ_{X_1})(f∘π_*) = a(X_1)f, which has the opposite sign. The resulting Leibniz rule is [X_1, fX_2] = f[X_1,X_2] - (a(X_1)f)X_2, contradicting condition 1 of Definition 2.3. The theorem can be repaired by defining a(X) = -Tπ_*♯(dλ_X), but as written the converse statement is not correct.","section":"Section 4, Theorem 4.3"},{"comment":"The Jacobi identity is not proved in the manuscript. The final step of the proof says that because the differentials of functions of the form λ_X + f∘π_* span T^*E_* and the Poisson bracket depends only on differentials, the Jacobi identity for all functions follows. This is not justified as written: the expression {f,{g,h}} involves derivatives of the Poisson tensor and can depend on second derivatives of f,g,h, so spanning of first differentials at each point does not by itself reduce the verification. A rigorous argument is needed, for example by proving that the Schouten bracket [Π,Π] vanishes by checking it on the spanning family of 1-forms and using the algebroid Jacobi identity, or by citing a precise statement in [CP12] that covers this predual setting. As written, the central existence claim of Theorem 3.2 is not fully supported.","section":"Section 3, Theorem 3.2"}],"minor_comments":[{"comment":"The sentence 'we obtain linear Banach Poisson structure on ℓ2×ℓ∞' should refer to the predual bundle, i.e. ℓ2×ℓ1, since the Poisson structure is constructed on E_*, not on E.","section":"Section 6, final paragraph"},{"comment":"The text 'from Proposition 3.2 we obtain a Poisson bracket' should read 'from Theorem 3.2', since the reference is to the main existence theorem of Section 3.","section":"Section 3, page 12"},{"comment":"The word 'precontangent' should be 'precotangent' in the sentence 'we discuss the case of weak symplectic structure on a precontangent bundle'.","section":"Introduction, page 2"},{"comment":"The proof uses convergence of the series Σ s_n e_n and continuity of the Lie bracket, but the Banach space topology on Γ(E) is never specified. Please state the functional-analytic assumptions on the space of sections that make the series argument valid.","section":"Theorem 2.4 proof"},{"comment":"The notation 's(m)∈(π_*^{-1}(m))^* = π^{-1}(m)⊂E' would be clearer if written as 'the fiber E_m of E', since π^{-1}(m) is a subset of the total space E, not a section.","section":"Section 4, Lemma 4.2"}],"recommendation":"major_revision","confidential_remarks":"The main construction in Section 3 appears to be essentially correct and is a reasonable contribution, but the proof of Jacobi identity needs to be made rigorous, and the sign error in Section 4 is a genuine flaw that must be fixed. The skeptical concern about a sign error in Lemma 3.3 does not survive a careful check: with the convention ♯(α)(β)={g_β,f_α} used in the paper, the sharp-map formula is consistent with (3.4). The self-citations are motivational and not problematic."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThe short version: this paper does what it claims. It constructs a genuine (full) linear Poisson bracket on the predual bundle E_* of a Banach Lie algebroid E, rather than the sub-Poisson structure on the dual bundle from [CP12, CP24]. The resulting conditions (3.6)--(3.7) for the bracket to define a Banach Poisson manifold are clean and match the Lie algebra case. The sign question that some readers will raise about Lemma 3.3 is not an error: with the convention ♯(α)(β)=Π(β,α) (so ♯(dλ_X)(df)={f,λ_X}), the claimed formula produces exactly the brackets in (3.4). The 'corrected' sharp map that a stress-test note proposes flips the Hamiltonian vector field and would give the wrong sign.\n\nWhat is genuinely new: the bijectivity of λ (Lemma 4.2) -- on the dual bundle only injectivity holds -- and the explicit characterization of when the predual bracket is a Banach Poisson manifold. The examples are honest and check the conditions; the trivial ℓ2×ℓ∞→ℓ2 example verifies A^*(ℓ2)⊂ℓ1, and the weak symplectic structure on the precotangent bundle is recovered as the special case.\n\nSoft spots: The Jacobi identity proof is delegated to [CP12] with a one-sentence spanning argument. The argument is essentially right (the trivector [Π,Π] is a pointwise tensor, so vanishing on the differentials of the basic functions at each point forces vanishing on all covectors), but the text should spell out that pointwise-tensor step. The proof in Theorem 4.3 that the anchor depends only on the value of X is terse; the separation into derivative and value parts works, but a reader has to work to see it. These are exposition gaps, not mathematical holes. The queer algebroid discussion is interesting and correctly sets the scope; the open problem for non-Schauder fibers is explicitly stated. Self-citations are motivational and not used to dodge proof burden.\n\nThe result is not revolutionary, but it is a useful and correct contribution to Banach Poisson geometry. I would send it to referees; with the proofs expanded in the places above, it can be accepted. I would cite it if I work on infinite-dimensional Lie–Poisson spaces.\n\nBest,\n[Your name]","headline":"A solid, correct adaptation of the Lie algebroid--Poisson correspondence to predual bundles; the apparent sign error in Lemma 3.3 is a pairing-order artifact, not a real flaw.","tokens_in":14071,"tokens_out":18165,"would_cite":true,"duration_ms":183913,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53D17","22E65","46T05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The predual bundle of a Banach Lie algebroid carries a canonical linear Poisson bracket, with Hamiltonian vector fields exactly under two dual-inclusion conditions.","keywords":["Banach Lie algebroid","predual bundle","linear Poisson structure","Banach Poisson manifold","queer Lie algebroid","precotangent bundle","weak symplectic form","Hamiltonian vector fields"],"falsifier":"Find a Banach Lie algebroid $E\\to M$ whose typical fiber has no Schauder basis and whose bracket $[\\cdot,\\cdot]$ depends on higher derivatives of sections, i.e. two local sections with the same first jet at some point have a bracket with different values there; for such a queer algebroid, $\\lambda_{[X,Y]}$ is not determined by $d\\lambda_X,d\\lambda_Y$, so the Poisson tensor in Theorem 3.2 is not well-defined and the proposed predual Poisson structure fails.","tokens_in":13008,"feed_emoji":"📐","tokens_out":6807,"duration_ms":65569,"temperature":0.7,"pith_summary":"This paper proves that a Banach Lie algebroid—an infinite-dimensional relative of a Lie algebra living over a manifold—has a natural Poisson geometry on its predual bundle, rather than on the dual bundle, which is usually too large in infinite dimensions. The construction works for all smooth functions, not just a restricted class, because the differentials of fiberwise-linear functions and base pullbacks span the full cotangent bundle of the predual. The paper also gives necessary and sufficient conditions, in terms of two dual-inclusion requirements, for this Poisson structure to have Hamiltonian vector fields for every smooth function, and it proves a converse: every linear Poisson structure of the right kind yields a Banach Lie algebroid on the dual bundle. This recovers the known Banach Lie–Poisson spaces as the special case of a Lie algebra, and it extends the weak-symplectic Poisson bracket on precotangent bundles to a genuine Poisson bracket.","feed_headline":"Predual bundle carries the Poisson structure of a Banach Lie algebroid","feed_subtitle":"Hamiltonian vector fields exist exactly when two dual-inclusion conditions hold; the bracket is defined on all smooth functions.","key_machinery":"The load-bearing mechanism is the Poisson tensor $\\Pi$ defined on two spanning families of functions, together with the observation that their differentials span $T^*E_*$ rather than only a characteristic subbundle. The tensor is fixed by $\\Pi(d\\lambda_X,d(f\\circ\\pi_*))=(a(X)f)\\circ\\pi_*$ and $\\Pi(d\\lambda_X,d\\lambda_Y)=\\lambda_{[X,Y]}$; non-queerness—the assumption that the Lie algebroid bracket depends only on first jets of sections—guarantees that $\\lambda_{[X,Y]}$ is a function of $d\\lambda_X,d\\lambda_Y$, so $\\Pi$ is well defined. The sharp map then has the explicit local form $\\sharp(m,\\varphi,\\mu,x)=(m,\\varphi,-a(x),a^*(\\mu)-(ad^m_x)^*\\varphi)$, and the two inclusions in Theorem 3.6 are precisely the conditions that make this map take values in $TE_*$.","core_discovery":"On the paper's own terms: for any localizable, non-queer Banach Lie algebroid $E\\to M$ that admits a predual bundle $E_*$, there is a canonical linear Poisson bracket on $E_*$ determined by the relations $\\{\\lambda_X,\\lambda_Y\\}=\\lambda_{[X,Y]}$, $\\{\\lambda_X,f\\circ\\pi_*\\}=(a(X)f)\\circ\\pi_*$, and $\\{f\\circ\\pi_*,g\\circ\\pi_*\\}=0$, where $\\lambda_X$ are the fiberwise-linear functions and $\\pi_*$ is the projection. The differentials of these functions span the whole cotangent bundle of $E_*$ (Proposition 3.1), so the bracket, defined first on this spanning family, extends to all smooth functions and satisfies the Jacobi identity. The predual $E_*$ becomes a Banach Poisson manifold in the sense of Definition 2.1 if and only if the dual anchor satisfies $a^*(T^*M)\\subset E_*$ and the duals of the bilinear maps $ad^m_x$ satisfy $(ad^m_x)^*(E_*)\\subset E_*$ for all $m$ and $x$. These two inclusions are exactly what makes the sharp map land in the tangent bundle, hence exactly the conditions under which Hamiltonian vector fields exist for every smooth function.","pith_inferences":["Going beyond the paper: the condition $a^*(T^*M)\\subset E_*$ is strong enough that even the identity anchor of a tangent bundle fails for non-reflexive manifolds; this suggests Hamiltonian vector fields on predual bundles will exist only for anchors whose dual is adapted to the chosen predual subbundle.","The paper leaves open whether queer Banach Lie algebroids exist on fibers without a Schauder basis; if one is ever constructed, Theorem 3.2's Poisson tensor would not be well defined on the predual, so the scope of this construction is tied to that open problem.","One testable extension would be to search systematically for anchors $A$ on trivial bundles $M\\times E$ that satisfy $A^*(T^*M)\\subset E_*$ and the adjoint-type inclusion; the paper's $\\ell^2$ example is one such instance, and many others could be built the same way.","This predual viewpoint may make Banach Poisson geometry accessible in settings where the cotangent space is hard to describe, such as spaces of compact operators as preduals of bounded operators on a Hilbert space."],"forward_implications":["For a Banach Lie algebra (base a single point), the theorem reduces to the Banach Lie–Poisson space criterion $ad^*_x E_*\\subset E_*$, recovering the known condition for the existence of Hamiltonian vector fields.","For $E=TM$ with a precotangent bundle $T_*M$, the canonical Poisson bracket extends the sub-Poisson bracket coming from the canonical weak symplectic form to all smooth functions on $T_*M$; in the non-reflexive case the two inclusions fail, so the precotangent bundle itself is not a Banach Poisson manifold.","For the trivial bundle $\\ell^2\\times\\ell^\\infty\\to\\ell^2$ with the specified anchor, the predual bundle $\\ell^2\\times\\ell^1$ is a Banach Poisson manifold if and only if $A^*(\\ell^2)\\subset\\ell^1$, giving a concrete infinite-dimensional example.","Conversely, any linear localizable non-queer Poisson bracket on $E_*$ satisfying the Definition 2.1 conditions yields a Banach Lie algebroid structure on $E$, closing the correspondence in the predual setting.","Because the spanning family is large enough, the resulting Poisson bracket is defined on all smooth functions of the predual bundle, not only on functions whose differentials lie in a characteristic subbundle as in the dual-bundle approach."],"supporting_citations":[{"why":"Supplies the Banach Lie–Poisson space criterion $ad^*_x E_*\\subset E_*$ that Theorem 3.6 generalizes from Lie algebras to algebroids.","marker":"[OR03]"},{"why":"Provides the dual-bundle sub-Poisson construction that the paper adapts, including the map $\\lambda$ whose injectivity, but not surjectivity, was established there.","marker":"[CP12]"},{"why":"Clarifies the non-queerness assumption and supplies the convenient-setting version of the Poisson–Lie algebroid correspondence used in the converse Theorem 4.3.","marker":"[CP24]"},{"why":"Constructs queer Poisson brackets depending on higher derivatives, motivating the exclusion of queer algebroids and the partial negative result in Theorem 2.4.","marker":"[BGT18]"},{"why":"Provides the framework of localizable Banach Lie algebroids and groupoids that justifies the sheaf-theoretic assumptions used throughout.","marker":"[BGJP19]"},{"why":"Contains the first-jet dependence statement whose Banach analogue under a Schauder-basis assumption rules out queer algebroids in Theorem 2.4.","marker":"[Mar02]"},{"why":"Original finite-dimensional construction of linear Poisson structures from Lie algebroids, which the present paper adapts to the predual setting.","marker":"[CDW87]"},{"why":"Finite-dimensional correspondence between linear Poisson structures and Lie algebroid structures, the model for both directions of the predual construction.","marker":"[Cou90]"}],"fun_headline_variants":["Poisson structure on predual bundle of Banach Lie algebroids","Banach Lie algebroid predual gets a linear Poisson bracket","Two inclusions guarantee Hamiltonian vectors on predual bundle","Alternative construction: Poisson structure on predual"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction assumes the Lie algebroid bracket is non-queer, meaning it depends only on first jets of sections, while the proof that queer brackets cannot exist is given only for fibers with a Schauder basis; a queer bracket on a fiber without one would make the Poisson tensor ill-defined, and the whole construction also presupposes a smooth predual bundle.","fun_headline_variants_meta":{"raw":{"variants":["Poisson structure on predual bundle of Banach Lie algebroids","Banach Lie algebroid predual gets a linear Poisson bracket","Two inclusions guarantee Hamiltonian vectors on predual bundle","Alternative construction: Poisson structure on predual"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00054,"raw_usage":{"total_tokens":2556,"prompt_tokens":877,"completion_tokens":1679,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":493,"completion_tokens_details":{"reasoning_tokens":1611}},"tokens_in":493,"tokens_out":1679,"duration_ms":12280,"temperature":1.0,"reasoning_tokens":1611,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:18:20.502263+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a Banach Lie algebroid $E\\to M$ whose typical fiber has no Schauder basis and whose bracket $[\\cdot,\\cdot]$ depends on higher derivatives of sections, i.e. two local sections with the same first jet at some point have a bracket with different values there; for such a queer algebroid, $\\lambda_{[X,Y]}$ is not determined by $d\\lambda_X,d\\lambda_Y$, so the Poisson tensor in Theorem 3.2 is not well-defined and the proposed predual Poisson structure fails.","supporting_citations":[],"review_version":1}