On the Shuffling Algorithm for the Aztec Nordenstam eno@kth.se - PowerPoint PPT Presentation

On the Shuffling Algorithm for the Aztec Diamond Eric On the Shuffling Algorithm for the Aztec Nordenstam eno@kth.se Diamond Background Shuffling algorithm Eric Nordenstam Warren’s Process eno@kth.se Aztec diamond point Universit´ e Catholique de Louvain, Belgium process Asymptotics Statcomb 09 Borodin & Ferrari

The Aztec Diamond On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Aztec diamonds of orders 1, 2, 3 and 4. Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

The Aztec Diamond On the Shuffling Algorithm for the Aztec Diamond Eric Aztec diamonds of orders 1, 2, 3 and 4. Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process The diamond of order n can be tiled in 2 n ( n +1) / 2 ways. Aztec diamond point Elkies et al, ’92 process Asymptotics Borodin & Ferrari

The Aztec Diamond On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

The Aztec Diamond y = 1 x = 16 On the y = 2 x = 15 Shuffling y = 3 x = 14 Algorithm for y = 4 x = 13 the Aztec y = 5 x = 12 Diamond y = 6 x = 11 y = 7 x = 10 Eric y = 8 x = 9 Nordenstam y = 9 x = 8 eno@kth.se y = 10 x = 7 y = 11 x = 6 Background y = 12 x = 5 y = 13 x = 4 Shuffling y = 14 x = 3 algorithm y = 15 x = 2 y = 16 x = 1 Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

GUE Minor Process On the Shuffling Let H be a large GUE matrix, i.e. a random matrix with Algorithm for the Aztec probability density Z − 1 e − Tr H 2 . Diamond Let H n = [ H i , j ] 1 ≤ i , j ≤ n be the n :th minor of H . Eric Nordenstam Let H n have eigenvalues λ n 1 , . . . , λ n eno@kth.se n Background Shuffling algorithm Warren’s Process Aztec diamond point process Theorem (Johansson&N ’06) Asymptotics Borodin & The Aztec diamond point process in a suitable rescaling Ferrari converges to the GUE minor process.

On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se 1 The shuffling algorithm Background 2 Warren’s Interlaced Brownian motions Shuffling algorithm 3 Asymptotics Warren’s Process 4 Borodin & Ferrari Aztec diamond point process Asymptotics Borodin & Ferrari

On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Three phases of the algorithm. Background 1 Delete Shuffling algorithm 2 Shuffle Warren’s 3 Create Process Aztec diamond point process Asymptotics Borodin & Ferrari

Delete On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling �→ algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Shuffle On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling �→ algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Create On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling �→ algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Particle dynamics On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

TASEP with step initial condition On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Particle dynamics On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Particle dynamics On the Shuffling Algorithm for the Aztec Diamond x 1 1 ( t ) = x 1 1 ( t − 1) + γ 1 1 ( t ) Eric Nordenstam x j 1 ( t ) = x j 1 ( t − 1) + γ j 1 ( t ) − 1 { x j 1 ( t − 1) + γ j 1 ( t ) = x j − 1 eno@kth.se ( t − 1) + 1 } 1 x j j ( t ) = x j j ( t − 1) + γ j j ( t ) + 1 { x j j ( t − 1) + γ j j ( t ) = x j − 1 Background j − 1 ( t − 1) } Shuffling x j i ( t ) = x j i ( t − 1) + γ j i ( t ) − 1 { x j i ( t − 1) + γ j i ( t ) = x j − 1 ( t − 1) + 1 } algorithm j Warren’s + 1 { x j i ( t − 1) + γ j i ( t ) = x j − 1 j − 1 ( t − 1) } . Process Aztec diamond point Here, γ j i ( t ) are i.i.d. fair coin tosses and initial conditions x j i ( j ) = i process for j = 1 , 2 , . . . and 1 ≤ i ≤ j . At each time t , Asymptotics Borodin & i ( t ) ≤ x j − 1 x j ( t ) ≤ x j i +1 ( t ) . Ferrari i

Aztec diamond particle dynamics On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam Perform substitution eno@kth.se Background X j i ( t ) = x j i ( t − j ) (1) Shuffling algorithm For all t , Warren’s Process i ( t ) ≤ X j − 1 X j ( t ) < X j i +1 ( t ) (2) i Aztec diamond point process Asymptotics Borodin & Ferrari

Aztec diamond particle dynamics On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Warren’s Process On the Shuffling Algorithm for the Aztec Diamond Eric Nordenstam eno@kth.se Background Shuffling algorithm Warren’s Process Aztec diamond point process Asymptotics Borodin & Ferrari

Transition Density for Dyson’s BM On the Shuffling Algorithm for the Aztec Let W n = { x ∈ R n : x 1 ≤ x 2 ≤ · · · ≤ x n } . For x , x ′ ∈ W n Diamond Eric Nordenstam ( x , x ′ ) = h n ( x ′ ) eno@kth.se p n , + � ϕ t ( x ′ � h n ( x ) det i − x j ) (3) t Background Shuffling where algorithm � h n ( x ) = ( x j − x i ) (4) Warren’s Process i < j Aztec diamond point and process 1 e − x 2 / 2 t √ ϕ t ( x ) = (5) Asymptotics 2 π t Borodin & Ferrari

On the Shuffling Algorithm for the Aztec Nordenstam eno@kth.se - PowerPoint PPT Presentation

On the Shuffling Algorithm for the Aztec Diamond Eric On the Shuffling Algorithm for the Aztec Nordenstam eno@kth.se Diamond Background Shuffling algorithm Eric Nordenstam Warrens Process eno@kth.se Aztec diamond point

Needs and Assets Assessment Prepared by Leanne Hathcock Aztec Public Library, Aztec, NM Aztec

Random matrices and Aztec diamonds Kurt Johansson Florence, May, 2015. Domino Tilings of the

New World Empires 1. Three American Empires in Comparison: Aztec, Incan, Spanish T

Two-periodic Aztec diamond Arno Kuijlaars (KU Leuven) joint work with Maurice Duits (KTH

Shuffling Cards via One-sided Transpositions Stephen Connor Joint work with Oliver Matheau-Raven

Permutations, Card Shuffling, and Representation Theory Franco Saliola, Universit du Qubec

Tiling Shuffling Phenomenon Tri Lai University of Nebraska Lincoln Lincoln, NE 68588 Dimers

1 2 THE RULES 5.1 Objectives Telpochcalli were Aztec schools attended by teenagers (from aged

Shuffling, Partitioning, and Closures Parallel Programming and Data Analysis Heather Miller What

Shuffling properties for products of random permutations Olivier Bernardi (MIT) Joint work with

Shuffling: A Lock Contention Aware Thread Scheduling Technique Kishore Pusukuri Multicores are

Amplification by Shuffling: From Local to Central Differential Privacy via Anonymity Vitaly

Shuffling the Deck Chairs Exploring the impacts of Gross versus Net Margin on Services Reporting

S5151 - Voting And Shuffling For Fewer Atomic Operations Elmar Westphal, Forschungszentrum

Aztec Municipal School District This is Our Story Kirk Carpenter - Superintendent Grant Banash

Analyzing the Shuffling Side-Channel Countermeasure for Lattice-Based Signatures Peter Pessl

COMP 3403 Algorithm Analysis Part 3 Chapters 6 7 Jim Diamond CAR 409 Jodrey School

Relativistic Effects Relativistic Bit . . . Can Keep Data Secret: Relativistic Bit . . . Why

Construction Algorithms for (Polynomial) Lattice Points Peter Kritzer Johann Radon Institute for

The Eight-Point Algorithm COMPSCI 527 Computer Vision COMPSCI 527 Computer Vision The

CS70: Today Euclids GCD algorithm. Multiplicative Inverse. (define (euclid x y) (if (= y 0)

Adversarial Search CE417: Introduction to Artificial Intelligence Sharif University of Technology

Outline ` Mining Sequential Patterns PrefixSpan: Mining Sequential Patterns Problem

CS 2550 / Spring 2006 Principles of Database Systems Undo/No-Redo No-Undo/Redo