{"id":"604ea872-995d-4fea-9184-e92ae07fe09c","arxiv_id":"2606.22301","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Closed-form expression for conditional mutual information in linear Gaussian DAGs, constructed from AD primitives, enables gradient-based optimization of multi-terminal wireless rate regions.","lead":"The paper gives a closed-form expression for conditional mutual information between groups of nodes in linear Gaussian networks, written as a log-determinant difference of Schur complements taken from covariances produced by one forward recursion. This expression is assembled from automatic-differentiation building blocks so that any objective built from such mutual informations becomes end-to-end differentiable with respect to network design parameters.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"The reader's weakest_assumption correctly isolates the sole modeling premise on which everything else rests. Because the abstract already states the premise explicitly and the subsequent algebraic steps are standard, the full-text description does not introduce any additional unsupported link that would alter the UNVERDICTED status or raise correctness_risk above medium.","tokens_in":1771,"tokens_out":341,"duration_ms":30031,"concrete_test":"Implement the K-recursion on a 4-node linear Gaussian DAG with known closed-form joint covariance; extract the three relevant 2x2 blocks for a chosen (A,B,C) triple, compute the claimed Schur-complement MI, and compare numerically to the direct formula ½ log(det(Σ_{AB|C})/det(Σ_{A|C}Σ_{B|C})) obtained from the full joint covariance matrix.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that any multi-terminal linear Gaussian network can be modeled exactly as a Gaussian-DAG whose node-pair covariances are produced by one forward K-recursion, after which I(V_A;V_B|V_C) is the standard log-det difference of two sub-block Schur complements. This modeling step is presented as a deliberate representational choice rather than a derived theorem; once the DAG and recursion are granted, the closed-form MI and its AD compatibility follow from elementary Gaussian identities (no hidden algebraic steps or unstated independence assumptions appear). The three listed experiments are consistent with this modeling scope (MIMO MAC, wiretap, multi-hop MAC) and do not require cycles or non-linearities.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript claims that multi-terminal linear Gaussian wireless networks can be modeled exactly as linear Gaussian DAGs, allowing node-pair covariances to be obtained via a single K-recursion forward pass; conditional mutual information I(V_A; V_B | V_C) is then given in closed form as the log-determinant difference of two sub-block Schur complements of the support covariance. This construction uses only AD primitives, so any differentiable function of finitely many such CMIs (including standard rate-region objectives) is end-to-end differentiable, with gradients obtained from a single reverse-mode AD sweep. The framework is illustrated on rate-region maximization for a two-user MIMO MAC, secure precoding for a MIMO wiretap channel, and rate-region optimization on a multi-hop MAC network.","tokens_in":1894,"tokens_out":511,"duration_ms":37793,"significance":"If the modeling choice and closed-form derivation hold, the work supplies a unified, parameter-free route to gradient-based physical-layer design for multi-terminal Gaussian channels that avoids problem-specific gradient derivations. Credit is due for the explicit reliance on standard covariance algebra and Schur-complement identities (no fitted parameters or circular definitions) together with the AD compatibility that directly yields Wirtinger gradients for linear and composite objectives. The three experiments stay within the stated scope of acyclic linear networks and are consistent with the claimed computational path.","major_comments":[{"comment":"The K-recursion forward pass (described in the modeling section preceding the closed-form MI expression) is load-bearing for the central claim that covariances are obtained exactly from one pass; an explicit algebraic verification or small worked example confirming the recursion for a multi-terminal DAG (e.g., the two-user MIMO MAC case) would allow direct checking for gaps or edge cases, consistent with the low soundness rating on the derivation.","section":"Modeling section (K-recursion forward pass)"}],"minor_comments":[{"comment":"Notation for the support covariance matrix and the sub-block Schur complements should be introduced with a short definitional paragraph before the log-det expression to improve readability for readers outside the immediate subfield.","section":"Notation and preliminaries"},{"comment":"Figure captions in the experimental sections would benefit from listing the precise channel dimensions, noise variances, and power constraints used, aiding reproducibility of the reported rate-region curves.","section":"Experiments"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the positive assessment and constructive comment. We address the major comment below and will incorporate the requested verification in the revision.","responses":[{"response":"We agree that an explicit worked example strengthens verifiability of the K-recursion. The recursion follows directly from the standard covariance propagation rule for linear Gaussian DAGs (each node's covariance is the sum of contributions from its parents plus independent noise), which is applied once per node in topological order. In the revised manuscript we will insert a short algebraic verification for the two-user MIMO MAC: we explicitly compute the node-pair covariances via the K-recursion for a 2x2 MIMO MAC with given channel matrices and noise variances, then confirm that the resulting joint covariance matrix matches the direct (non-recursive) block-matrix expression obtained from the linear model. This addition will be placed immediately after the general K-recursion statement and before the closed-form CMI expression.","revision_made":"yes","referee_comment":"[Modeling section (K-recursion forward pass)] The K-recursion forward pass (described in the modeling section preceding the closed-form MI expression) is load-bearing for the central claim that covariances are obtained exactly from one pass; an explicit algebraic verification or small worked example confirming the recursion for a multi-terminal DAG (e.g., the two-user MIMO MAC case) would allow direct checking for gaps or edge cases, consistent with the low soundness rating on the derivation."}],"tokens_in":1447,"tokens_out":317,"duration_ms":22162,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The paper's core move is to treat the multi-terminal linear Gaussian network as a Gaussian DAG, run a single K-recursion to obtain all node-pair covariances, then express I(V_A; V_B | V_C) directly as the log-det difference of two sub-block Schur complements of the joint covariance. Once that is granted, any differentiable function of finitely many such terms becomes end-to-end differentiable, so weighted sum-rate, secrecy rate, or composite objectives can be optimized by projected gradient steps without writing a new gradient expression for each topology.\n\nThat construction is new in the wireless information-theory literature. Prior work either left the gradients to be derived case-by-case or used black-box estimators; here the expression is algebraic and sits inside an AD graph. The three reported experiments (two-user MIMO MAC, MIMO wiretap, and a larger multi-hop MAC) are consistent with the modeling scope and show that the same code path handles different rate-region problems.\n\nThe main limitation is the modeling premise itself: the network must be exactly representable as a linear Gaussian DAG whose covariances come from one forward K-recursion. Topologies with cycles or essential nonlinearities fall outside the stated guarantee, though the paper does not claim to cover them. The derivation steps themselves appear to be standard covariance identities, so the algebraic risk looks low once the DAG representation is accepted.\n\nThe work is aimed at people doing physical-layer optimization on Gaussian multi-user channels who already use gradient methods. It supplies a reusable primitive rather than a single new rate region. A serious referee should see it because the primitive is cleanly stated, the experiments are on-point, and the scope is explicit.","headline":"Gives a closed-form, AD-compatible expression for conditional MI on Gaussian DAGs via Schur complements after one covariance recursion, which removes the need for hand-derived gradients on standard multi-terminal rate objectives.","tokens_in":2369,"tokens_out":422,"would_cite":true,"duration_ms":24137,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Conditional mutual information in linear Gaussian networks is a log-determinant difference of Schur complements.","keywords":["conditional mutual information","Gaussian DAG","Schur complement","differentiable optimization","multi-terminal channels","automatic differentiation","rate region","wireless networks"],"falsifier":"For a two-user MIMO multiple-access channel with known closed-form rate region, compute the formula's value for a chosen covariance and compare it to the standard mutual-information expression evaluated on the same covariance.","tokens_in":2664,"feed_emoji":"📡","tokens_out":714,"duration_ms":30882,"temperature":0.7,"pith_summary":"The paper derives a closed-form expression for any conditional mutual information I(V_A; V_B | V_C) among groups of nodes in a multi-terminal linear Gaussian wireless network. By representing the network as a linear Gaussian directed acyclic graph, all required node-pair covariances follow from a single forward recursion, after which the mutual information is obtained directly as a difference of two log-determinants of sub-block Schur complements. The entire construction uses only automatic-differentiation primitives, so any objective composed of finitely many such terms, including weighted sum-rates, secrecy rates, and full rate-region functions, is end-to-end differentiable with respect to all controllable parameters. A single reverse-mode sweep then supplies the gradient for all parameters at once.","feed_headline":"One recursion yields closed-form conditional MI for Gaussian networks","feed_subtitle":"The resulting expression is built from AD primitives, so rate-region and secrecy objectives become end-to-end differentiable.","key_machinery":"The K-recursion forward pass on a linear Gaussian DAG that produces all node-pair covariances, from which conditional MI is extracted as the log-determinant difference of two Schur complements.","core_discovery":"We obtain I(V_A;V_B | V_C) in closed form: from the node-pair covariances produced by one K-recursion forward pass, it is a log-determinant difference of two sub-block Schur complements of the support covariance. The construction is built entirely from automatic-differentiation primitives, so any differentiable function of finitely many conditional MIs is end-to-end differentiable in the design parameters; this broad class includes linear objectives (weighted sum-rate, secrecy), the rate functions of standard multi-terminal rate regions, and non-linear composites of these.","pith_inferences":["The same covariance recursion might support joint optimization of discrete coding choices if they can be relaxed into the differentiable graph.","Feedback or time-varying channels could be handled by unfolding the recursion over multiple time steps while preserving the DAG structure.","Similar closed-form extractions may exist for other information quantities such as directed information in Gaussian settings."],"forward_implications":["Weighted sum-rate and secrecy-rate objectives become directly optimizable by projected gradient methods on the design parameters.","Rate-region maximization for MIMO multiple-access, broadcast, and interference channels can use the same gradient procedure.","The same construction applies without change to larger multi-hop networks.","Any non-linear but differentiable composite of several conditional MIs remains end-to-end differentiable."],"fun_headline_variants":["K-recursion produces closed-form conditional MI via Schur complements","Log-determinant Schur differences give differentiable conditional MI","Gaussian-DAG covariances enable closed-form conditional MI computation","Conditional MI built from AD primitives for multi-terminal networks","Single forward pass computes Schur-based conditional MI in Gaussian-DAG"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The multi-terminal network can be represented exactly as a linear Gaussian directed acyclic graph whose node-pair covariances are obtained by a single K-recursion forward pass.","fun_headline_variants_meta":{"raw":{"variants":["K-recursion produces closed-form conditional MI via Schur complements","Log-determinant Schur differences give differentiable conditional MI","Gaussian-DAG covariances enable closed-form conditional MI computation","Conditional MI built from AD primitives for multi-terminal networks","Single forward pass computes Schur-based conditional MI in Gaussian-DAG"]},"model":"grok-4.3","cost_usd":0.004931,"raw_usage":{"total_tokens":2454,"prompt_tokens":748,"num_sources_used":0,"completion_tokens":81,"cost_in_usd_ticks":49312000,"prompt_tokens_details":{"text_tokens":748,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1625,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":748,"tokens_out":81,"duration_ms":18002,"temperature":1.0,"reasoning_tokens":1625,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-26T10:11:00.974039+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"For a two-user MIMO multiple-access channel with known closed-form rate region, compute the formula's value for a chosen covariance and compare it to the standard mutual-information expression evaluated on the same covariance.","supporting_citations":[],"review_version":1}