{
 "artifacts": null,
 "category": "training",
 "description": "Replace the raw HJB residual loss of a neural PDE solver with a parametrix-preconditioned fixed-point target. At each local space-time patch, analytically propagate terminal values and source terms through a Gaussian kernel whose covariance uses a frozen diffusion matrix, while asking the network to learn only the variable-coefficient correction. This should reduce the burden on the network to represent stiff high-frequency diffusion dynamics and improve short-horizon convergence.",
 "formulas_latex": [
  "$$\\partial_{t}V(t;x)+\\inf_{u}\\left\\{L^{u}V(t;x)+\\ell(t;x)+u^{\\top}Ru\\right\\}=0,\\qquad V(T;x)=\\varphi(x),$$",
  "$$L^{u}v(t;x)=\\bigl(f(t;x)+g(t;x)u\\bigr)\\cdot\\nabla_{x}v(t;x)+\\frac{1}{2}\\,\\sigma(t;x)\\sigma(t;x)^{\\top}:D_{x}^{2}v(t;x),$$",
  "$$\\widehat{P_h^{A_0}q}(\\xi)=\\exp\\!\\left(-\\frac12h\\,\\xi^{\\top}A_0\\xi\\right)\\hat q(\\xi),\\qquad A_0=\\sigma(\\tau,x_0)\\sigma(\\tau,x_0)^{\\top},$$",
  "$$V(t)=P_{T-t}^{A_0}\\varphi+\\int_t^T P_{s-t}^{A_0}\\!\\left[\\ell(s,\\cdot)+H(s,\\cdot,\\nabla V(s,\\cdot))+\\frac12\\bigl(A(s,\\cdot)-A_0\\bigr):D^2V(s,\\cdot)\\right]ds.$$"
 ],
 "id": 2976,
 "implementation": "Integrate the method into a neural HJB solver at the training-target and residual-evaluation stage, not into the network architecture itself. Use an MLP $V_\\theta(t,x)$ and a backward time grid $t_0\u003c\\cdots\u003ct_N=T$. Partition sampled states into local patches with centers $(\\tau_j,x_j)$; estimate $A_j=\\sigma(\\tau_j,x_j)\\sigma(\\tau_j,x_j)^\\top$, symmetrize it, and add $\\epsilon I$ if needed. For each patch and timestep, compute the network quantities $p=\\nabla_xV_\\theta(t_{n+1},x)$ and $B=D_x^2V_\\theta(t_{n+1},x)$ using autodiff. Evaluate the Hamiltonian $H=\\min_u[(f+gu)^\\top p+u^\\top Ru]$; for quadratic control with invertible $R$, use $u^*=-\\frac12R^{-1}g^\\top p$ and substitute it into $H$. Form the source correction $q=\\ell+H+\\frac12(A(t_{n+1},x)-A_j):B$. Apply the frozen Gaussian propagator $P_{\\Delta t}^{A_j}$ to $q$ by either Monte Carlo samples $x'=x+L_j\\sqrt{\\Delta t}\\,z$, where $L_jL_j^\\top=A_j$, averaging $q(t_{n+1},x')$, or a truncated cosine/Fourier feature approximation using the multiplier $\\exp[-\\frac12\\Delta t\\xi^\\top A_j\\xi]$. Construct the backward target $y=P_{\\Delta t}^{A_j}V_\\theta(t_{n+1},\\cdot)+\\Delta tP_{\\Delta t}^{A_j}q$ and minimize $\\|V_\\theta(t_n,x)-y\\|^2$ over sampled points, optionally adding a small raw PDE residual term. Run one Picard sweep per epoch, detach the target for stability, and refresh patch centers every few epochs. The paper supplies the frozen-diffusion/variable-coefficient decomposition; Monte Carlo error, patch assignment, and timestep size are empirical choices. First test on a 10-dimensional linear-quadratic diffusion-control problem with known Riccati solution, comparing a vanilla PINN residual baseline against the same MLP with raw backward targets. Measure value-function relative error, gradient error, loss decrease per optimizer step, and sensitivity to $\\Delta t$. Success is faster decrease and lower error at equal network evaluations, especially when $A(x)$ varies strongly across the state space.",
 "math_summary": "The paper studies the HJB equation $\\partial_t V+\\inf_u\\{L^uV+\\ell+u^\\top Ru\\}=0$ with terminal condition $V(T,x)=\\varphi(x)$, where $L^uV=(f+gu)\\cdot\\nabla V+\\frac12 A:D^2V$ and $A=\\sigma\\sigma^\\top$ is the state-dependent diffusion covariance. The transferable parametrix operation freezes $A$ at a patch center $(\\tau,x_0)$, writing $A_0=A(\\tau,x_0)$, and uses the constant-coefficient Gaussian semigroup with Fourier multiplier $\\widehat{P_h^{A_0}q}(\\xi)=\\exp[-\\frac12h\\,\\xi^\\top A_0\\xi]\\hat q(\\xi)$ for step size $h\u003e0$. The variable-coefficient remainder is $\\frac12(A(t,x)-A_0):D^2V$. With $H(t,x,p)=\\inf_u\\{(f(t,x)+g(t,x)u)\\cdot p+u^\\top Ru\\}$, the adapted backward fixed-point equation is $V(t)=P_{T-t}^{A_0}\\varphi+\\int_t^T P_{s-t}^{A_0}[\\ell(s,\\cdot)+H(s,\\cdot,\\nabla V(s,\\cdot))+\\frac12(A(s,\\cdot)-A_0):D^2V(s,\\cdot)]\\,ds$. Here $P_h^{A_0}$ performs exact diffusion smoothing, $D^2V$ is the Hessian, and the correction vanishes when the diffusion is locally constant.",
 "math_tags": [
  "pde",
  "stochastic-processes",
  "harmonic-analysis",
  "numerical-analysis",
  "dynamical-systems"
 ],
 "ml_areas": [
  "training",
  "optimizer",
  "world-model"
 ],
 "paper": {
  "arxiv_id": "2609.00529",
  "arxiv_url": "https://arxiv.org/abs/2609.00529",
  "summary_what_math_gives_to_ml": "The transferable asset is the paper's parametrix treatment of state-dependent diffusion: instead of approximating a variable-coefficient parabolic operator directly, freeze the diffusion locally, apply an exactly computable Gaussian propagator, and represent the coefficient mismatch as a correction. This suggests a preconditioned neural PDE solver in which the network learns only the residual left after analytic diffusion smoothing, rather than the full HJB solution. The same construction can be implemented as a short-horizon Picard or policy iteration with frozen Gaussian convolutions and automatic-differentiation correction terms. The approach is most promising for high-dimensional stochastic-control surrogates, where grid methods fail and ordinary PINN residual optimization is poorly conditioned.",
  "title": "Finite-Horizon Hamilton--Jacobi--Bellman Equations with State-Dependent Diffusion in Spectral Barron Spaces",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 6,
  "novelty": 7,
  "usefulness": 7
 },
 "solves": [
  "stability",
  "speedup",
  "accuracy"
 ],
 "title": "Frozen-Diffusion Parametrix Preconditioner",
 "url": "https://synthcore.org/idea/2976/frozen-diffusion-parametrix-preconditioner",
 "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
   }
  }
 }
}