{
 "artifacts": null,
 "category": "geometry",
 "description": "Regularize a neural representation so that no one-dimensional projection places too much probability mass inside a narrow interval. This transfers the paper's uniform tube estimate into an anti-collapse constraint, making representations robust to adversarial directions and preventing hidden features from becoming effectively low-dimensional.",
 "formulas_latex": [
  "$$\\dim(\\mu):=\\inf\\{\\dim_H A:\\mu(A)\u003e0\\},\\qquad \\dim(\\pi_u\\mu)=\\dim(\\mu)\\ \\text{for every }u\\in S^1.$$",
  "$$\\mu\\big(\\{z\\in\\mathbb{R}^2:\\operatorname{dist}(z,L)\\leq w\\}\\big)\\leq Cw^s\\quad\\text{for every line }L,\\ 0\u003cw\\leq 1.$$",
  "$$(\\pi_u\\mu)([a-w,a+w])\\leq C'(2w)^s,\\qquad \\pi_u(z)=u^\\top z.$$",
  "$$\\mathcal{L}_{\\mathrm{tube}}=\\frac{1}{|\\mathcal{U}|}\\sum_{u\\in\\mathcal{U}}\\sum_{w\\in\\mathcal{W}}\\left[\\log q_{u,w}^{\\max}-\\log C-s\\log(2w)\\right]_+^2,\\quad q_{u,w}^{\\max}=\\max_a\\frac{1}{B}\\sum_{i=1}^B\\sigma\\!\\left(\\frac{w-|u^\\top z_i-a|}{\\tau}\\right).$$"
 ],
 "id": 2837,
 "implementation": "Integrate this after the encoder and before the task head. Use a two-dimensional projection head \\(z_i=g(f_\\theta(x_i))\\); retain the task loss, and optimize \\(\\mathcal{L}=\\mathcal{L}_{task}+\\lambda\\mathcal{L}_{tube}\\). At each step, sample 32--128 unit directions \\(u\\) uniformly on the circle, use logarithmically spaced widths \\(\\mathcal{W}=\\{2^{-2},2^{-3},\\ldots,2^{-7}\\}\\) after batchwise normalization of z, and estimate the worst interval center by evaluating centers at the projected batch values plus a fixed grid. Pseudocode: compute z; for each u compute r=u^T z; for each w and center a compute the soft count q using the sigmoid in the displayed formula; take the maximum over a; accumulate the hinge-square penalty; backpropagate task loss plus penalty. The paper supplies the target power law and its all-directions interpretation; the batch counts, constant C, smoothing temperature \\(\\tau\\), and exponent s are empirical design choices. Start with s=0.7, \\(\\lambda\\in\\{10^{-3},10^{-2},10^{-1}\\}\\), and \\(\\tau=0.05w\\). The first cheap test is a two-layer MLP on MNIST or CIFAR-10 with cross-entropy, comparing the regularizer against no regularizer and standard variance normalization. Probe 512 directions, including adversarially optimized directions, and report the maximum projected interval mass versus w. The prediction is a log-log slope at least s-0.1 over the unsaturated scales for the regularized model, while the baseline has at least one direction with a substantially smaller slope; increasing \\(\\lambda\\) should reduce the worst-direction mass monotonically until task accuracy begins to degrade. Also test robustness after adding a small perturbation to inputs: the regularized representation should show a smaller increase in worst-direction concentration.",
 "math_summary": "Let \\(\\mu\\) be the distribution of a two-dimensional hidden representation \\(z=f_\\theta(x)\\), let \\(u\\in S^1\\) be a unit projection direction, and let \\(\\pi_u(z)=u^\\top z\\). The paper's mechanism is the uniform tube/Frostman estimate \\(\\mu(\\{z:\\operatorname{dist}(z,L)\\leq w\\})\\leq Cw^s\\) for every line \\(L\\), width \\(0\u003cw\\leq 1\\), and exponent \\(0\u003cs\u003c1\\). Equivalently, every projected measure obeys \\((\\pi_u\\mu)([a-w,a+w])\\leq C'(2w)^s\\) for every \\(u\\), center \\(a\\), and \\(w\\). For a minibatch \\(z_i\\), estimate the left side with a differentiable kernel count and penalize violations of \\(\\log q(u,a,w)\\leq \\log C+s\\log(2w)\\), where \\(q\\) is the empirical projected small-ball mass. The property doing the work is uniformity over \\(u\\) and \\(a\\), not merely good behavior for randomly sampled projections.",
 "math_tags": [
  "geometry",
  "measure-theory",
  "harmonic-analysis",
  "dynamical-systems"
 ],
 "ml_areas": [
  "embedding",
  "regularization",
  "mlp"
 ],
 "paper": {
  "arxiv_id": "2608.29569",
  "arxiv_url": "https://arxiv.org/abs/2608.29569",
  "summary_what_math_gives_to_ml": "The paper constructs ergodic ×p-invariant measures whose mass in every planar tube of width w is O(w^s), so no line projection loses dimension, even when the projected iterated-function system has exact overlaps. The transferable mechanism is a uniform Frostman/anti-concentration condition over all projection directions, rather than an average-case random-projection guarantee. In neural networks, this can become a representation regularizer that penalizes excessive concentration of hidden embeddings inside thin slabs or intervals after any one-dimensional projection. The main falsifiable prediction is that the worst-direction projected small-ball probability follows a power law with exponent s, while unconstrained representations exhibit a direction and scale range with a smaller exponent.",
  "title": "Ergodic $\\times p$-invariant measures on $\\mathbb{T}^2$ with no dimension dropping projections",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 4,
  "novelty": 7,
  "usefulness": 6
 },
 "solves": [
  "generalization",
  "stability"
 ],
 "title": "All-Direction Frostman Representation",
 "url": "https://synthcore.org/idea/2837/all-direction-frostman-representation",
 "verification": {
  "peer_reviewed": false,
  "status": "unverified",
  "status_label": "Unverified",
  "verdict_source": "deterministic test code (paired-seed permutation statistics)"
 }
}