{
 "artifacts": null,
 "category": "stability",
 "description": "Use a Sylvester equation as an implicit parameterization of a recurrent or state-space transition matrix, while regularizing the actual solution X rather than only the coefficient matrices A and B. The key mechanism is to suppress small singular values of X and near-resonant eigenvalue pairs, which otherwise amplify selected components of the learned transition.",
 "formulas_latex": [
  "$$AX-XB=C$$",
  "$$(I\\otimes A-B^{T}\\otimes I)\\operatorname{vec}(X)=\\operatorname{vec}(C)$$",
  "$$X_{ij}=\\frac{C_{ij}}{\\lambda_i-\\mu_j}\\qquad\\text{when }A=\\operatorname{diag}(\\lambda_i),\\;B=\\operatorname{diag}(\\mu_j)$$",
  "$$\\mathcal{L}_{\\mathrm{cond}}=\\log\\left(\\frac{\\sigma_{\\max}(X)+\\varepsilon}{\\sigma_{\\min}(X)+\\varepsilon}\\right)+\\rho\\sum_{i,j}\\frac{|C'_{ij}|^2}{|\\lambda_i-\\mu_j|^2+\\varepsilon^2}$$"
 ],
 "id": 3160,
 "implementation": "Integrate this into an implicit recurrent or state-space layer. Store trainable A, B, and C, solve AX-XB=C with a differentiable Sylvester solver, and use X as the transition in h_{t+1}=phi(X h_t+W u_t+b). A minimal training loop is: (1) construct A, B, and C, optionally restricting A and B to diagonal or small block-diagonal form; (2) compute X with a batched Sylvester solve; (3) estimate sigma_max(X) by power iteration on X transpose times X; (4) estimate sigma_min(X) by power iteration on the inverse or by a differentiable small-matrix SVD; (5) add lambda_cond times the smoothed log condition number to the task loss; (6) for diagonal or block-diagonal coefficients, compute eigenvalue gaps and add the resonance term in the displayed formula; (7) backpropagate through the solve. The mathematical quantities supplied by the paper are the Sylvester structure and the small-denominator risk. Singular values, eigenvalues, and coefficients are estimated numerically from the current batch-independent parameters. First test on sequential MNIST or a delayed-copy task with a small state size, comparing an unconstrained recurrent matrix, an unregularized Sylvester layer, and the conditioned layer at matched parameter counts. Record long-horizon accuracy, gradient-norm variance, hidden-state norms, solver residuals, and loss spikes. Success means fewer exploding trajectories, smoother gradients, and higher delayed-copy accuracy at equal parameter count without materially increasing per-step solve cost.",
 "math_summary": "The Sylvester equation is AX-XB=C, where A is an m by m matrix, B is an n by n matrix, C and X are m by n matrices, and X is the learned transition or a factor used to construct it. Uniqueness holds when A and B have no common eigenvalue. Vectorization gives (I tensor A minus B transpose tensor I) vec(X)=vec(C), where vec stacks matrix columns and I is the identity. The paper shows that X may be arbitrarily ill-conditioned even when A, B, C, and the Kronecker operator are well-conditioned. For diagonal A=diag(lambda_i) and B=diag(mu_j), each entry satisfies X_ij=C_ij/(lambda_i-mu_j), so small eigenvalue gaps create amplification. The condition number is kappa_2(X)=sigma_max(X)/sigma_min(X), where sigma_max and sigma_min are the largest and smallest singular values. A practical penalty is a smoothed log condition number plus a resonance penalty. For diagonalizable A and B, C_prime denotes C expressed in their eigenvector bases.",
 "math_tags": [
  "linear-algebra",
  "spectral-theory",
  "numerical-analysis"
 ],
 "ml_areas": [
  "rnn",
  "ssm",
  "regularization"
 ],
 "paper": {
  "arxiv_id": "2609.04050",
  "arxiv_url": "https://arxiv.org/abs/2609.04050",
  "summary_what_math_gives_to_ml": "The paper identifies a failure mode relevant to implicit, recurrent, and state-space neural layers: the Sylvester solution X can become arbitrarily ill-conditioned even when A, B, C, and the associated Kronecker operator are individually well-conditioned. Thus, monitoring coefficient norms, solver residuals, or spectral separation alone does not guarantee stable learned representations or gradients. A practical transfer is to parameterize a neural transition through a Sylvester equation and explicitly regularize the conditioning of its solution, especially near eigenvalue resonances that create small-denominator amplification.",
  "title": "Conditioning of solutions to the Sylvester equation",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 5,
  "novelty": 7,
  "usefulness": 6
 },
 "solves": [
  "stability",
  "accuracy"
 ],
 "title": "Conditioned Sylvester Neural Layer",
 "url": "https://synthcore.org/idea/3160/conditioned-sylvester-neural-layer",
 "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
   }
  }
 }
}