{
 "artifacts": null,
 "category": "architecture",
 "description": "Replace repeated high-dimensional quadratic interactions in a neural operator or world model with a reduced latent quadratic map whose basis-pair interactions are precomputed offline. The online computation becomes a small polynomial in latent coefficients, preserving quadratic interaction structure while avoiding repeated full-resolution contractions.",
 "formulas_latex": [
  "$$Q(f,f)(\\mathbf{x},\\mathbf{v})=\\int_{\\mathbb{R}^{d}}\\int_{\\mathbb{S}^{d-1}}B(\\mathbf{v}-\\mathbf{v}_{*},\\sigma)[f(\\mathbf{x},\\mathbf{v}_{*}^{\\prime})f(\\mathbf{x},\\mathbf{v}^{\\prime})-f(\\mathbf{x},\\mathbf{v}_{*})f(\\mathbf{x},\\mathbf{v})]\\\\,\\mathrm{d}{\\sigma}\\\\,\\mathrm{d}{\\mathbf{v}_{*}}.$$",
  "$$f_{\\mathrm{rb}}=\\sum_{i=1}^{n}c_i u_i,\\qquad Q(f_{\\mathrm{rb}},f_{\\mathrm{rb}})=\\sum_{i=1}^{n}\\sum_{j=1}^{n}c_i c_j Q(u_i,u_j).$$",
  "$$G_{kij}=\\langle u_k,Q(u_i,u_j)\\rangle,\\qquad q_k(c)=\\sum_{i=1}^{n}\\sum_{j=1}^{n}G_{kij}c_i c_j.$$",
  "$$G_{kij}\\approx\\sum_{r=1}^{R}A_{kir}B_{rij},\\qquad q_k(c)\\approx\\sum_{r=1}^{R}\\sum_{i,j}A_{kir}B_{rij}c_i c_j.$$"
 ],
 "id": 3066,
 "implementation": "(1) Integration point: add a quadratic latent operator after the encoder of a neural operator, graph simulator, or world model. The encoder maps an input field or graph state x to c in R^n. Replace a dense high-resolution pairwise interaction or MLP block with q(c), concatenate [c,q(c)], and pass the result to the decoder or next residual block. This changes both training and inference computation.\n\n(2) Pseudocode:\n```text\n# Offline\nfor i in range(n):\n  for j in range(n):\n    h = expensive_quadratic_operator(u[i], u[j])\n    for k in range(n):\n      G[k,i,j] = inner_product(u[k], h)\noptionally fit low-rank factors A,B to G\n\n# Online\nc = encoder(x)\nif low_rank:\n  t[r] = sum(i,j, B[r,i,j] * c[i] * c[j])\n  q[k] = sum(r, A[k,r] * t[r])\nelse:\n  q[k] = sum(i,j, G[k,i,j] * c[i] * c[j])\ny = decoder(concat(c, q))\n```\nUse symmetry when justified, storing only i\u003c=j and doubling off-diagonal terms. Initialize the quadratic block and its decoder projection near zero so a pretrained baseline is preserved initially. (3) The basis-pair expansion and offline interaction tensor are taken from the paper; the learned encoder, decoder, basis, and factor rank R are engineering choices. Estimate R by randomized SVD or Tucker decomposition of G, retaining 99% of interaction energy. (4) First experiment: train a small Fourier neural operator or graph simulator on a 2D parametric advection-reaction benchmark. Compare a standard MLP interaction, a dense quadratic interaction, and the precomputed reduced operator at matched latent width. Measure forward latency, peak memory, rollout error, and inverse-problem optimization time. Success is at least 2x lower inference cost or memory at matched rollout error, with stable gradients.",
 "math_summary": "The paper's collision operator Q(f,f) is quadratic in f. For a reduced representation f_rb=sum_{i=1}^n c_i u_i, where u_i are fixed basis functions and c_i are latent coefficients, bilinearity gives Q(f_rb,f_rb)=sum_{i,j} c_i c_j Q(u_i,u_j). Define G_{kij}=\u003cu_k,Q(u_i,u_j)\u003e, where k is the reduced output coordinate and \u003c.,.\u003e is the selected inner product. The reduced output is q_k(c)=sum_{i,j}G_{kij}c_i c_j. All expensive pairwise operator evaluations occur offline. Online inference only contracts the stored tensor G with c tensor c. A separable or low-rank approximation G_{kij} approximately equal to sum_{r=1}^R A_{kir}B_{rij} can reduce storage and computation; R is the retained factorization rank.",
 "math_tags": [
  "tensor-decomposition",
  "linear-algebra",
  "pde",
  "approximation-theory"
 ],
 "ml_areas": [
  "world-model",
  "graph-nn",
  "inference-speedup"
 ],
 "paper": {
  "arxiv_id": "2609.02578",
  "arxiv_url": "https://arxiv.org/abs/2609.02578",
  "summary_what_math_gives_to_ml": "The paper combines residual-based greedy reduced-basis construction, offline/online separation of a quadratic nonlocal operator, and constrained residual minimization in reduced coordinates. The transferable asset is the representation of a parameterized quadratic map as a small polynomial in latent coefficients with precomputed basis-pair interactions. This can produce compact quadratic latent operators for neural operators, graph simulators, and world models, while residual-greedy sampling can adaptively allocate expensive training data to difficult regions of parameter space. The strongest first experiments are on parametric PDE surrogates, where full-resolution interaction costs and repeated inverse-problem solves are measurable.",
  "title": "Reduced order model for parametric Boltzmann equation and its application to inverse problems",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 5,
  "novelty": 6,
  "usefulness": 6
 },
 "solves": [
  "speedup",
  "memory",
  "scalability"
 ],
 "title": "Offline quadratic latent operator",
 "url": "https://synthcore.org/idea/3066/offline-quadratic-latent-operator",
 "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
   }
  }
 }
}