• Corpus ID: 244798713

On the Gaussian surface area of spectrahedra

@article{Arunachalam2021OnTG,
  title={On the Gaussian surface area of spectrahedra},
  author={Srinivasan Arunachalam and Oded Regev and Penghui Yao},
  journal={Electron. Colloquium Comput. Complex.},
  year={2021},
  volume={28},
  pages={178}
}
We show that for sufficiently large n ≥ 1 and d = Cn 3 / 4 for some universal constant C > 0, a random spectrahedron with matrices drawn from Gaussian orthogonal ensemble has Gaussian surface area Θ( n 1 / 8 ) with high probability. 

Positive spectrahedra: invariance principles and pseudorandom generators

TLDR
An invariance principle for positive spectrahedra is proved via the well-known Lindeberg method, an upper bound on noise sensitivity and a Littlewood-Offord theorem are proved and applications are given for constructing PRGs forpositive spectahedra, learning theory, discrepancy sets for positiveSpectrahedRA (over the Boolean cube) andPRGs for intersections of structured polynomial threshold functions.

References

SHOWING 1-10 OF 17 REFERENCES

Topics in random matrix theory, volume 132 of Graduate Studies in Mathematics

  • American Mathematical Society,
  • 2012

Positive spectrahedra: invariance principles and pseudorandom generators

TLDR
An invariance principle for positive spectrahedra is proved via the well-known Lindeberg method, an upper bound on noise sensitivity and a Littlewood-Offord theorem are proved and applications are given for constructing PRGs forpositive spectahedra, learning theory, discrepancy sets for positiveSpectrahedRA (over the Boolean cube) andPRGs for intersections of structured polynomial threshold functions.

Adaptive estimation of a quadratic functional by model selection

We consider the problem of estimating ∥s∥ 2 when s belongs to some separable Hilbert space and one observes the Gaussian process Y(t) = (s, t) + σ L(t), for all t ∈ H, where L is some Gaussian

Simple and efficient pseudorandom generators from gaussian processes

TLDR
A deterministic algorithm that can approximately count satisfying assignments of intersections of k general LTFs over {0, 1}n in time poly(n) · 2poly(log k,1/ε) for all values of k.

The reverse isoperimetric problem for Gaussian measure

  • K. Ball
  • Mathematics
    Discret. Comput. Geom.
  • 1993
AbstractIt is shown that ifg is the standard Gaussian density on ℝn andC is a convex body in ℝn % MathType!MTEF!2!1!+-% feaafiart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn%

Fooling Intersections of Low-Weight Halfspaces

  • R. ServedioLi-Yang Tan
  • Mathematics, Computer Science
    2017 IEEE 58th Annual Symposium on Foundations of Computer Science (FOCS)
  • 2017
TLDR
The analysis of the PRG fuses techniques from two different lines of work on unconditional pseudorandomness for different kinds of Boolean functions and introduces new coupling-based ingredients into the standard Lindeberg method for establishing quantitative central limit theorems and associated pseudor randomness results.

A simple proof of Stirling's formula for the gamma function

where and the notation means that as . C = 2π f (n) ∼ g(n) f (n) / g(n) → 1 n → ∞ A great deal has been written about Stirling's formula. At this point I will just mention David Fowler's Gazette