~/talk

Maximal Inequalities on the Hypercube

links:none on file
slides:none on file

Proves a dimension-independent L2 bound for the maximal average over Hamming spheres on {0,1}^n, so the number of vertices ruined by an adversary that colours an ε-fraction is O(ε) rather than depending on n. Proof combines Stein-style reductions, spectral estimates on the sphere operators via Kravchuk polynomials, a comparison to senate operators, and domination by a noise semigroup (Poisson clocks) plus Cauchy–Schwarz; yields a constant operator norm and therefore the claimed combinatorial corollary.

AI-written summary · about

  1. 0:00Intro
  2. 0:11Talk outline
  3. 0:44Motivation 1: An Election Interpretation
  4. 4:13Motivation 2: UGC or "Can We Hope for Better Approximation Algorithms in P?"
  5. 5:45Unique Games - Unique Label Cover Problem
  6. 7:46Unique Games Conjecture
  7. 11:12Where to begin if we want to refute UGC?
  8. 12:41UGC and the Spectrum of General Graphs
  9. 15:33An Algorithm
  10. 18:56The Adversary
  11. 19:56The Question
  12. 21:19The Conjecture
  13. 22:30The Challenge
  14. 29:13Convert to a problem in Analysis
  15. 31:49Maximal Operator
  16. 33:51Maximal Inequalities: a little history
  17. 38:52Step 2 of Proof
  18. 41:32Step 1+ Step 2
  19. 42:02Step 1 of Proof
  20. 45:03Spherical-Mean Maximal Inequality: method
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.