{
 "artifacts": null,
 "category": "architecture",
 "description": "Replace selected residual, recurrent, or state-space blocks by modules whose input-output Jacobians satisfy an IODP inequality throughout a prescribed activation domain. The constraint controls incremental amplification between two trajectories without requiring either trajectory to remain near one fixed equilibrium, so it should improve robustness to changing contexts and prevent exploding long-horizon sensitivities.",
 "formulas_latex": [
  "$$-\\sigma I+\\frac{J(u)+J(u)^{\\mathsf T}}{2}-\\rho J(u)^{\\mathsf T}J(u)\\succeq0,\\qquad J(u)=\\frac{\\partial h_\\theta(u)}{\\partial u}.$$",
  "$$\\begin{bmatrix}PA+A^{\\mathsf T}P+\\epsilon I\u0026PB\\\\B^{\\mathsf T}P\u00260\\end{bmatrix}-\\begin{bmatrix}0\u0026I\\\\C\u0026D\\end{bmatrix}^{\\mathsf T}\\begin{bmatrix}-\\sigma I\u0026I/2\\\\I/2\u0026-\\rho I\\end{bmatrix}\\begin{bmatrix}0\u0026I\\\\C\u0026D\\end{bmatrix}\\preceq0.$$",
  "$$M^{\\mathsf T}PM-P\\preceq-\\alpha P,\\qquad M=\\frac{\\partial z_{k+1}}{\\partial z_k},\\quad P\\succ0,\\quad 0\u003c\\alpha\u003c1.$$",
  "$$\\|\\delta z_k\\|_P\\le(1-\\alpha)^{k/2}\\|\\delta z_0\\|_P,\\qquad \\|v\\|_P^2=v^{\\mathsf T}Pv.$$"
 ],
 "id": 2975,
 "implementation": "Integration point: begin with an MLP residual block or a small recurrent cell, because its input and state Jacobians can be computed with automatic differentiation. Parameterize P=LLᵀ+δI, where L is learned and δ=10⁻⁴, and choose target indices σ,ρ and margin α\u003e0. For each minibatch, sample actual activations plus perturbed points within the intended domain, compute J by Jacobian-vector products, and form the symmetric violation matrix S=-σI+(J+Jᵀ)/2-ρJᵀJ. Penalize positive eigenvalues using L_IODP=softplus(λmax(-S)+m)², where m is a safety margin. For a residual block, compute the exact Jacobian M of the implemented update, or use M=I+η∂rθ/∂z, then add softplus(λmax(P⁻¹ᐟ²(MᵀPM-P+αP)P⁻¹ᐟ²))². Pseudocode is: sample (z,u); compute J or M; estimate the largest eigenvalue with 5 to 10 power iterations; add the violation penalty to task loss; backpropagate; periodically evaluate dense-domain violations and empirical trajectory-distance ratios. The paper supplies the local matrix inequality and compositional interpretation; the neural implementation estimates worst-case violations from minibatch and adversarially perturbed samples, so certification is empirical unless interval or spectral bounds are added. First cheap experiment: train a 2-layer residual MLP on MNIST classification at depths 10, 50, and 100, comparing unconstrained training with IODP-constrained training. Run identical-input trajectory tests from pairs of perturbed hidden states while sweeping η. The predicted signature is that constrained hidden-distance ratios remain below (1-α)¹ᐟ² per step, while the unconstrained model crosses ratio one when the largest generalized eigenvalue of MᵀPM-P reaches zero. The observed divergence boundary in η should track the LMI boundary within approximately 20 percent; increasing α should lower the stable η ceiling but produce exponential perturbation decay.",
 "math_summary": "For a static neural map y=h(u) with Jacobian J(u)=∂h/∂u, the paper's IODP condition with indices (σ,ρ) is -σI+(J+Jᵀ)/2-ρJᵀJ⪰0 for every u in the certified domain. Here I is the input-dimension identity, σ is the passivity index, and ρ is the output-differential index. For a dynamic neural block ẋ=f(x,u), y=h(x,u), let A=∂f/∂x, B=∂f/∂u, C=∂h/∂x, and D=∂h/∂u. A positive-definite matrix P=Pᵀ≻0 and ε\u003e0 certify the local differential storage V=δxᵀPδx/2 when the paper's LMI holds: [PA+AᵀP+εI, PB; BᵀP, 0] - [0,I; C,D]ᵀ[-σI,I/2; I/2,-ρI][0,I; C,D]⪯0. The first term supplies differential decay through ε; the second supplies incremental input-output dissipation. For a discrete residual block zₖ₊₁=zₖ+ηrθ(zₖ,uₖ), use its sampled Jacobian M=∂zₖ₊₁/∂zₖ and require MᵀPM-P⪯-αP with P≻0 and α\u003e0; this predicts trajectory-distance decay by at least (1-α)ᵏ for identical inputs.",
 "math_tags": [
  "control-theory",
  "dynamical-systems",
  "optimization",
  "linear-algebra",
  "spectral-theory"
 ],
 "ml_areas": [
  "mlp",
  "rnn",
  "ssm",
  "training-dynamics"
 ],
 "paper": {
  "arxiv_id": "2609.00497",
  "arxiv_url": "https://arxiv.org/abs/2609.00497",
  "summary_what_math_gives_to_ml": "The paper provides a transferable compositional stability mechanism: input-output differential passivity (IODP), certified using local Jacobian inequalities and a quadratic differential storage matrix. Unlike equilibrium-specific Lyapunov analysis, IODP is defined over a domain of state-input pairs and therefore certifies an entire equilibrium set, making it suitable for neural modules whose operating point changes during training or inference. The most direct transfer is a passivity-constrained neural block: estimate or bound its Jacobians, enforce the local matrix inequality with a differentiable penalty or projection, and compose blocks while monitoring the resulting contraction or incremental-gain margin.",
  "title": "Decentralized and Equilibrium-Set-Oriented Stability Analysis and Control for Power Systems",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 6,
  "novelty": 6,
  "usefulness": 8
 },
 "solves": [
  "stability",
  "generalization",
  "accuracy"
 ],
 "title": "Differentially Passive Neural Blocks",
 "url": "https://synthcore.org/idea/2975/differentially-passive-neural-blocks",
 "verification": {
  "peer_reviewed": false,
  "status": "unverified",
  "status_label": "Unverified",
  "verdict_source": "deterministic test code (paired-seed permutation statistics)",
  "verification_axes": {
   "benchmark_mechanism": {
    "confirmed": null,
    "tested": false
   },
   "practical_benchmark": {
    "beats_baseline": null,
    "tested": false
   },
   "toy_mechanism_gate": {
    "confirmed": null,
    "tested": false
   }
  }
 }
}