{
 "artifacts": null,
 "category": "architecture",
 "description": "Partition neural modules into two empirically identified reliability or noise classes and restrict their communication graph to a two-block stochastic block model. Allocate a fixed connectivity budget across within-class and cross-class edges using a water-filling update that favors block pairs producing the largest increase in validation utility. The resulting layer is sparse and modular, with a testable prediction that optimal connectivity concentrates on a few block pairs rather than remaining uniformly distributed.",
 "formulas_latex": [
  "$$\\mathcal{W}_{\\rho}=\\left\\{W:[0,1]^2\\to[0,1]\\mid W(x,y)=W(y,x),\\ \\int_0^1\\!\\int_0^1W(x,y)\\,dx\\,dy=\\rho\\right\\}.$$",
  "$$\\sum_{i,j=1}^{2}\\alpha_i\\alpha_jw_{ij}=\\rho,\\qquad 0\\leq w_{ij}\\leq1,\\qquad w_{12}=w_{21}.$$",
  "$$q_i=\\sigma\\!\\left(\\beta_i\\sum_{j=1}^{2}\\alpha_jw_{ij}\\left[U_{ij}(1,q_j)-U_{ij}(0,q_j)\\right]\\right),\\qquad \\sigma(z)=\\frac{1}{1+e^{-z}}.$$",
  "$$w_{ij}\\leftarrow\\operatorname{clip}_{[0,1]}\\!\\left(w_{ij}+\\gamma\\left[g_{ij}-\\lambda\\right]\\right),\\qquad g_{ij}=\\frac{\\partial F}{\\partial w_{ij}},\\qquad \\sum_{i,j}\\alpha_i\\alpha_jw_{ij}=\\rho.$$"
 ],
 "id": 2775,
 "implementation": "1. Integration point: apply the construction to a transformer attention layer, an MoE expert communication layer, or a graph-neural-network message-passing layer. Partition modules into two classes using an observable noise score, such as routing entropy, gradient variance, activation corruption sensitivity, or disagreement under data augmentation. Replace dense module-to-module communication by a two-block soft mask with probabilities w_11, w_12, and w_22. At inference, sample a fixed sparse mask or retain the highest-scoring edges subject to the same density. 2. Pseudocode: initialize the three block probabilities at target density rho; run minibatches with differentiable or straight-through Bernoulli masks; estimate F and each marginal g_ij by backpropagation or finite differences; update w_ij with clip(w_ij+gamma*(g_ij-lambda)); find lambda by bisection so the weighted density constraint is exact; periodically resample masks and recompute class scores. 3. Taken from the paper are the graphon relaxation, fixed-density constraint, SBM reduction, logit response analogy, and water-filling/KKT rule. Estimated empirically are alpha_i, beta_i, utilities, and gradients. 4. First experiment: train a small transformer on CIFAR-10 patches or WikiText-2 and compare dense attention, random block sparsity, unconstrained learned sparsity, and water-filled sparsity at equal FLOPs. Sweep rho and record validation loss, gradient variance, and communication disagreement. Prediction: at least one of w_11, w_12, or w_22 reaches near-zero or near-one, and performance versus rho has a kink when a block enters or leaves saturation. The selected active block ranking should agree with measured marginal utilities in at least 80% of runs; absence of saturation or kinks falsifies the mechanism.",
 "math_summary": "A graphon is a symmetric measurable connection function W(x,y) in [0,1] with fixed density rho, meaning the integral of W over the unit square equals rho. For two module classes with fractions alpha_1 and alpha_2=1-alpha_1, use block probabilities w_ij in [0,1], with w_12=w_21, and impose the density constraint sum over i,j of alpha_i alpha_j w_ij=rho. Let q_i be the probability that class i produces a useful or cooperative representation, beta_i be its inverse noise or rationality parameter, and U_ij(a,b) be the measured utility when classes i and j use actions a and b. The logit dynamics are q_i=sigma(beta_i Delta U_i), where Delta U_i=sum_j alpha_j w_ij[U_ij(1,q_j)-U_ij(0,q_j)] and sigma(z)=(1+exp(-z))^(-1). For a neural objective F(w), water filling updates each block according to its marginal utility g_ij=partial F/partial w_ij while a Lagrange multiplier lambda preserves the density constraint. At an interior optimum, all active blocks have equal marginal utility; blocks at zero or one satisfy the corresponding KKT inequality.",
 "math_tags": [
  "graph-theory",
  "optimization",
  "dynamical-systems",
  "probability",
  "statistics"
 ],
 "ml_areas": [
  "moe-routing",
  "transformer",
  "graph-nn",
  "optimizer"
 ],
 "paper": {
  "arxiv_id": "2608.27851",
  "arxiv_url": "https://arxiv.org/abs/2608.27851",
  "summary_what_math_gives_to_ml": "The paper provides a constructive graphon-design mechanism: for agents with a bimodal rationality profile following logit learning in a stag-hunt coordination game, an optimal interaction graphon can be searched for within the much smaller family of stochastic block models (SBMs). The fixed edge-density constraint converts topology design into a continuous allocation problem over block-to-block connection probabilities, solved locally by a water-filling procedure that saturates the most valuable block pairs first. This can transfer to modular neural systems by designing block-sparse attention, recurrent communication, or MoE expert-interaction graphs according to module reliability and noise classes rather than learning an unconstrained dense topology.",
  "title": "Graphon Design for Human-Machine Coordination under Bounded Rationality: Optimality of Stochastic Block Models",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 5,
  "novelty": 7,
  "usefulness": 6
 },
 "solves": [
  "stability",
  "scalability",
  "accuracy"
 ],
 "title": "Water-Filled Block-Sparse Neural Connectivity",
 "url": "https://synthcore.org/idea/2775/water-filled-block-sparse-neural-connectivity",
 "verification": {
  "peer_reviewed": false,
  "status": "unverified",
  "status_label": "Unverified",
  "verdict_source": "deterministic test code (paired-seed permutation statistics)"
 }
}