{
 "artifacts": null,
 "category": "architecture",
 "description": "Use fixed or lightly trainable rank-one Gaussian PSD measurements as a compact feature layer for representations whose useful information is contained in a low-rank Gram or covariance matrix. The nonnegative intensity coordinates preserve the geometry of low-rank PSD inputs while using only \\(m\\) scalar features instead of a full \\(d\\times d\\) matrix.",
 "formulas_latex": [
  "$$A_i=\\bm a_i\\bm a_i^*,\\qquad \\langle A_i,xx^*\\rangle=\\operatorname{Tr}(\\bm a_i\\bm a_i^*xx^*)=|\\bm a_i^*\\bm x|^2.$$",
  "$$X=HH^*,\\qquad \\Phi_{\\mathcal A}(X)_i=\\langle a_i a_i^*,HH^*\\rangle=\\|a_i^*H\\|_2^2\\ge0.$$",
  "$$\\beta_{\\mathcal A}(r)=\\frac{U_{\\mathcal A}(r)}{L_{\\mathcal A}(r)},\\qquad \\mathcal A=\\{a_i a_i^*\\}_{i=1}^{m},\\quad a_i\\overset{\\mathrm{i.i.d.}}{\\sim}\\mathcal N(0,I_d/d).$$"
 ],
 "id": 59,
 "implementation": "Integrate the layer after a token encoder and before a memory-heavy attention, pooling, or retrieval stage. Given token features \\(H\\in\\mathbb R^{d\\times s}\\), sample one fixed matrix \\(A\\in\\mathbb R^{m\\times d}\\) at initialization with entries \\(A_{ik}\\sim\\mathcal N(0,1/d)\\), and output \\(z\\in\\mathbb R^m\\) with `z=(A@H).square().mean(dim=1)`. This is exactly the rank-one PSD map because each row \\(a_i^\\top\\) defines \\(A_i=a_i a_i^\\top\\); do not form the \\(d\\times d\\) Gram matrix. Optionally add a trainable positive scale \\(g_i=\\operatorname{softplus}(u_i)\\) and use `z=g*z`, while keeping Gaussian directions frozen to isolate the mathematical design. For a decoder needing signed information, concatenate a conventional low-dimensional linear projection or apply this layer only to covariance-invariant branches, since intensity features do not preserve the sign of \\(H\\). Start with \\(m\\in\\{d/4,d/2,d\\}\\), rank-controlled synthetic data \\(H=UV^\\top\\) with rank \\(r\\in\\{2,4,8\\}\\), and a CIFAR-10 or small-language-model feature-compression task. Compare against dense random projection, learned linear projection, and explicit Gram-vectorization followed by PCA. Record activation memory, wall-clock throughput, reconstruction or task error, and pairwise distortion \\(q(X,Y)=\\|\\Phi(X)-\\Phi(Y)\\|_1/\\|X-Y\\|_F\\). The target is comparable task accuracy with reduced covariance storage and stable distortion while \\(r\\) remains below the design rank.",
 "math_summary": "The paper's rank-one construction sets \\(A_i=a_i a_i^*\\), with sensing vector \\(a_i\\in\\mathbb H^n\\), and obtains \\(\\langle A_i,xx^*\\rangle=|a_i^*x|^2\\). More generally, for \\(X=HH^*\\), the same measurement is \\(\\langle a_i a_i^*,HH^*\\rangle=\\|a_i^*H\\|_2^2\\), which is nonnegative and computable without materializing \\(X\\). The paper studies \\(\\beta_{\\mathcal A}(r)=U_{\\mathcal A}(r)/L_{\\mathcal A}(r)\\) and reports that random rank-one Gaussian measurements are asymptotically sharp for the universal lower bounds, making them near-optimal for low-rank PSD stability. Here \\(d\\) is feature dimension, \\(s\\) is the number of tokens or samples in \\(H\\), \\(m\\) is compressed width, and \\(a_i\\sim\\mathcal N(0,I_d/d)\\) are independent Gaussian sensing vectors.",
 "math_tags": [
  "linear-algebra",
  "random-matrix",
  "metric-geometry"
 ],
 "ml_areas": [
  "embedding",
  "memory",
  "inference-speedup",
  "architecture"
 ],
 "paper": {
  "arxiv_id": "2608.22418",
  "arxiv_url": "https://arxiv.org/abs/2608.22418",
  "summary_what_math_gives_to_ml": "The paper turns positive-semidefinite sensing into a quantitative stability problem: the relevant object is the ratio between the largest and smallest measurement distortion over pairs of low-rank PSD matrices. Its transferable asset is a principled way to design nonnegative, low-dimensional measurements of covariance-like neural representations, rather than treating compression as an unconstrained linear projection. Rank-one Gaussian measurements are shown to be asymptotically optimal for this condition-number objective, suggesting both a useful initialization and a fixed architecture for stable low-rank feature bottlenecks. A practical neural-network adaptation is to measure PSD feature matrices with intensity features and explicitly monitor or regularize empirical lower and upper Lipschitz constants.",
  "title": "Optimal Condition Numbers in Low-Rank Positive Semidefinite Matrix Sensing",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 3,
  "novelty": 7,
  "usefulness": 6
 },
 "solves": [
  "memory",
  "stability",
  "scalability"
 ],
 "title": "Rank-one Gaussian intensity feature layer",
 "url": "https://synthcore.org/idea/59/rank-one-gaussian-intensity-feature-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
   }
  }
 }
}