1 / 33

Discrepancy Minimization by Walking on the Edges

Discrepancy Minimization by Walking on the Edges. Raghu Meka (IAS & DIMACS) Shachar Lovett (IAS). Discrepancy. Subsets Color with or - to minimize imbalance. 1 2 3 4 5. 1 2 3 4 5. 3. 1. 1. 0. 1. Discrepancy Examples. Fundamental combinatorial concept.

swain
Download Presentation

Discrepancy Minimization by Walking on the Edges

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Discrepancy Minimization by Walking on the Edges Raghu Meka (IAS & DIMACS) Shachar Lovett (IAS)

  2. Discrepancy • Subsets • Color with or -to minimize imbalance 1 2 3 4 5 123 45 3 1 1 0 1

  3. Discrepancy Examples • Fundamental combinatorial concept • Arithmetic Progressions Roth 64: Matousek, Spencer 96:

  4. Discrepancy Examples • Fundamental combinatorial concept • Halfspaces Alexander 90: Matousek 95:

  5. Discrepancy Examples • Fundamental combinatorial concept • Axis-aligned boxes Beck 81: Srinivasan 97:

  6. Why Discrepancy? Complexity theory Communication Complexity Computational Geometry Pseudorandomness Many more!

  7. Spencer’s Six Sigma Theorem Spencer 85: System with n sets has discrepancy at most . “Six standard deviations suffice” • Central result in discrepancy theory. • Beats random: • Tight: Hadamard.

  8. A Conjecture and a Disproof Conjecture (Alon, Spencer): No efficient algorithm can find one. Bansal 10: Can efficiently get discrepancy . Spencer 85: System with n sets has discrepancy at most . • Non-constructive pigeon-hole proof

  9. This Work Main: Can efficiently find a coloring with discrepancy New elemantary constructive proof of Spencer’s result • Truly constructive • Algorithmic partial coloring lemma • Extends to other settings EDGE-WALK: New algorithmic tool

  10. Beck-Fiala Setting Each element occurs in at most sets 1 2 3 4 5 Beck-Fiala 80: Srinivasan 97: (log n)) (non-algorithmic) Banszczyk 98: (non-algorithmic)

  11. Beck-Fiala Setting Thm: Can efficiently find a coloring with discrepancy Each element occurs in at most sets Matches Bansal 10

  12. Outline Partial coloring Method EDGE-WALK: Geometric picture Analysis of algorithm

  13. Partial Coloring Method • Beck 80: find partial assignment with zeros 1-111-1 1-10 0 0 11 0 -1

  14. Partial Coloring Method Lemma: Can do this in randomized time. Input: Output: • Focus on m = n case.

  15. Outline Partial coloring Method EDGE-WALK: Geometric picture Analysis of algorithm

  16. Discrepancy: Geometric View • Subsets • Color with or -to minimize imbalance 123 45 3 1 1 0 1

  17. Discrepancy: Geometric View • Vectors • Want 123 45

  18. Discrepancy: Geometric View • Vectors • Want Polytope view used earlier by Gluskin’ 88. Goal: Find non-zero lattice point inside

  19. Edge-Walk Claim: Will find good partial coloring. • Start at origin • Gaussian walk until you hit a face • Gaussian walk within the face Goal: Find non-zero lattice point in

  20. Edge-Walk: Algorithm Gaussian random walk in subspaces • Subspace V, rate • Gaussian walk in V Standard normal in V: Orthonormal basis change

  21. Edge-Walk Algorithm Discretization issues: hitting faces • Might not hit face • Slack: face hit if close to it.

  22. Edge-Walk: Algorithm • Input: Vectors • Parameters: For Cube faces nearly hit by . Disc. faces nearly hit by . Subspace orthogonal to

  23. Edge-Walk: Intuition Discrepancy faces much farther than cube’s Hit cube more often! 100 1

  24. Outline Partial coloring Method EDGE-WALK: Geometric picture Analysis of algorithm

  25. Edge-Walk: Analysis Lem: For with prob 0.1 and

  26. Edge-Walk Analysis • Claim 1: Never cross polytope. • Claim 2: Number of disc. faces hit . • Win-Win: Hit many cube faces or grow

  27. Edge-Walk Analysis Claim 1: Never cross polytope • Must make a big jump • Unlikely:

  28. Edge-Walk Analysis • Claim 2: Number of disc. faces hit • Small progress each step • Martingale tail bound: • . • Linearity of expectation 100

  29. Edge-Walk Analysis • Claim 3: Hit many cube faces - • norm grows dimension of subspace • Final dimension small:

  30. Main Partial Coloring Lemma Th: Given thresholds Can find with 1. 2. Algorithmic partial coloring lemma

  31. Summary Beck-Fiala setting similar Spencer’s Theorem Edge-Walk: Algorithmic partial coloring lemma Recurseon unfixed variables

  32. Open Problems • Some promise: our PCL “stronger” than Beck’s Q: Beck-Fiala Conjecture 81: Discrepancy for degree t. Q: Other applications? General IP’s, Minkowski’s theorem? • Constructive version of Banszczyk’s bound?

  33. Thank you

More Related