All work and no play makes Jack enumerate maps Michael La Croix PhD - - PowerPoint PPT Presentation

all work and no play makes jack enumerate maps
SMART_READER_LITE
LIVE PREVIEW

All work and no play makes Jack enumerate maps Michael La Croix PhD - - PowerPoint PPT Presentation

All work and no play makes Jack enumerate maps Michael La Croix PhD University of Waterloo February 15, 2011 Outline Combinatorial Enumeration 1 Graphs, Maps, and Surfaces 2 Rooted Maps and Flags 3 Quantum gravity and the q -Conjecture 4


slide-1
SLIDE 1

All work and no play makes Jack enumerate maps

Michael La Croix PhD

University of Waterloo

February 15, 2011

slide-2
SLIDE 2

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 1 / 38

slide-3
SLIDE 3

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-4
SLIDE 4

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Combinatorics

The study of how simple sets can be combined to create more complicated sets.

Enumeration

Systematic counting.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-5
SLIDE 5

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Combinatorics

The study of how simple sets can be combined to create more complicated sets.

Enumeration

Systematic counting.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-6
SLIDE 6

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Combinatorics

The study of how simple sets can be combined to create more complicated sets.

Enumeration

Systematic counting.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-7
SLIDE 7

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Combinatorics

The study of how simple sets can be combined to create more complicated sets.

Enumeration

Systematic counting.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-8
SLIDE 8

My Perspective

I’m a mathematician. I study combinatorial enumeration. I think a lot of problems are best understood via pictures.

Combinatorics

The study of how simple sets can be combined to create more complicated sets.

Enumeration

Systematic counting.

Example

How many ways are there to arrange 7 black and 3 white marbles in a row?

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 2 / 38

slide-9
SLIDE 9

Generating Series (or Partition Functions)

A generating series is an algebraic tool for recording a sequences of

  • numbers. Using such tools, counting problems become algebra problems.

Example

If an,k is the number of ways to arrange n black and k white marbles, then

  • n,k≥0

an,kxnyk =

  • n,k≥0

n + k k

  • xnyk =

1 1 − x − y

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 3 / 38

slide-10
SLIDE 10

Generating Series (or Partition Functions)

A generating series is an algebraic tool for recording a sequences of

  • numbers. Using such tools, counting problems become algebra problems.

Example

If an,k is the number of ways to arrange n black and k white marbles, then

  • n,k≥0

an,kxnyk =

  • n,k≥0

n + k k

  • xnyk =

1 1 − x − y

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 3 / 38

slide-11
SLIDE 11

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-12
SLIDE 12

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-13
SLIDE 13

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-14
SLIDE 14

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-15
SLIDE 15

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-16
SLIDE 16

Why Should We Count Things?

It could be the key to computing probabilities. It is a first step in listing all the possible cases for a case analysis.

By counting we may implicitly describe how to list the objects. A count lets us check that our list is complete.

The behaviour of a sequence of numbers can describe physical properties.

This is the idea behind Statistical mechanics, a branch of physics that qualitatively models phase transitions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-17
SLIDE 17

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 4 / 38

slide-18
SLIDE 18

What is mathematics ?

Mathematics is a way of abstractly studying relationships between objects. From the webcomic Spiked Math at spikedmath.com

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 5 / 38

slide-19
SLIDE 19

Extra Information

By studying the abstraction, we can more easily recognize what extra information is essential to the structure of a problem.

Example (A Graph)

The graph K4,1.

Example (Chemistry)

Hydrogen Carbon Chlorine

The molecule chloromethane.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 6 / 38

slide-20
SLIDE 20

Extra Information

By studying the abstraction, we can more easily recognize what extra information is essential to the structure of a problem.

Example (A Graph)

The graph K4,1.

Example (Chemistry)

Hydrogen Carbon Chlorine

The molecule chloromethane.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 6 / 38

slide-21
SLIDE 21

Labelled Vertices

With labelled graphs, we can distinguish between some kinds of isomers.

Example (Isomers of dichloroethene) 1,1-dichloroethene 1,2-dichloroethene

The 1, 1 and 1, 2 isomers of C2H2Cl2 are represented by different graphs.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 7 / 38

slide-22
SLIDE 22

Embeddings

For other isomers, we actually need to draw the graphs.

Example (cis-trans isomers) cis-1,2-dichloroethene trans-1,2-dichloroethene

cis-trans isomers have the same graph, but different embeddings.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 8 / 38

slide-23
SLIDE 23

Embeddings

For other isomers, we actually need to draw the graphs.

Example (cis-trans isomers) cis-1,2-dichloroethene trans-1,2-dichloroethene

cis-trans isomers have the same graph, but different embeddings. Conclusion: Sometimes the way a graph is drawn is as important as the graph itself.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 8 / 38

slide-24
SLIDE 24

Three Utilities

GAS SEWERS WATER

A classical challenge is how to place conduits so that three utilities can be connected to three houses. This should be done so that no conduits cross.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-25
SLIDE 25

Three Utilities

Together, the houses, utilities, and conduits define a graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-26
SLIDE 26

Three Utilities

We want to embed the graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-27
SLIDE 27

Three Utilities

After a lot of effort, we con- clude that the problem cannot be solved as it appears to be stated.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-28
SLIDE 28

Three Utilities

After a lot of effort, we con- clude that the problem cannot be solved as it appears to be stated. This is actually a classic result (Kuratowski’s Theorem). The given graph is one of two ob- stacles to being able to draw a graph on the plane.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-29
SLIDE 29

Three Utilities

Solving the problems relies on finding a loop-hole in its state- ment.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-30
SLIDE 30

Three Utilities

Solving the problems relies on finding a loop-hole in its state- ment. One solution is that, as stated, the problem does not say that the houses are on a plane. We can draw them on a torus.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 9 / 38

slide-31
SLIDE 31

Representing Surfaces

The torus (or any surface) can be represented schematically in terms of the surgery required to stitch it together from a rubber sheet.

=

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 10 / 38

slide-32
SLIDE 32

Representing Surfaces

The torus (or any surface) can be represented schematically in terms of the surgery required to stitch it together from a rubber sheet.

Example

=

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 10 / 38

slide-33
SLIDE 33

Other Surfaces are Also Obtained by Surgery

= =

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 11 / 38

slide-34
SLIDE 34

Graphs, Surfaces, and Maps

Definition

A surface is a compact 2-manifold without boundary.

Definition

A graph is a finite set of vertices together with a finite set of edges, such that each edge is associated with either one or two vertices.

Definition

A map is a 2-cell embedding of a graph in a surface.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 12 / 38

slide-35
SLIDE 35

Graphs, Surfaces, and Maps

Definition

A surface is a compact 2-manifold without boundary.

Definition

A graph is a finite set of vertices together with a finite set of edges, such that each edge is associated with either one or two vertices.

Definition

A map is a 2-cell embedding of a graph in a surface.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 12 / 38

slide-36
SLIDE 36

Graphs, Surfaces, and Maps

Definition

A surface is a compact 2-manifold without boundary.

Definition

A graph is a finite set of vertices together with a finite set of edges, such that each edge is associated with either one or two vertices.

Definition

A map is a 2-cell embedding of a graph in a surface.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 12 / 38

slide-37
SLIDE 37

Maps and Faces

Once a graph is drawn, the unused portion of the paper is split into faces.

f1 f2 f3

A map is a graph together with an embedding in a surfaces. It is defined by its vertices, edges, and faces.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 13 / 38

slide-38
SLIDE 38

Maps and Faces

Once a graph is drawn, the unused portion of the paper is split into faces. For symmetry, the

  • uter

face is thought of as part of a sphere. A map is a graph together with an embedding in a surfaces. It is defined by its vertices, edges, and faces.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 13 / 38

slide-39
SLIDE 39

Tiling the Representation

The faces of a map can be made more evident by tessellating the tile that represents the surface.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 14 / 38

slide-40
SLIDE 40

Tiling the Representation

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 14 / 38

slide-41
SLIDE 41

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 14 / 38

slide-42
SLIDE 42

Ribbon Graphs

Example

The homeomorphism class of an embedding is determined by a neighbourhood of the graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 15 / 38

slide-43
SLIDE 43

Ribbon Graphs

Example

Neighbourhoods of vertices and edges can be replaced by discs and ribbons to form a ribbon graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 15 / 38

slide-44
SLIDE 44

Flags

Example

The boundaries of ribbons determine flags.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 15 / 38

slide-45
SLIDE 45

Flags

Example

The boundaries of ribbons determine flags, and these can be associated with quarter edges.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 15 / 38

slide-46
SLIDE 46

Rooted Maps

Definition

A rooted map is a map together with a distinguished orbit of flags under the action of its automorphism group.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 16 / 38

slide-47
SLIDE 47

Rooted Maps

Definition

A rooted map is a map together with a distinguished orbit of flags under the action of its automorphism group.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 16 / 38

slide-48
SLIDE 48

Rooted Maps

Definition

A rooted map is a map together with a distinguished orbit of flags under the action of its automorphism group.

Example

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 16 / 38

slide-49
SLIDE 49

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 16 / 38

slide-50
SLIDE 50

2-dimensional Quantum Gravity

Two models of 2-dimensional quantum gravity are analyzed by enumerating rooted orientable maps. The Penner Model involves all smooth maps. φ − 4 model involves only 4-regular maps.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 17 / 38

slide-51
SLIDE 51

2-dimensional Quantum Gravity

Two models of 2-dimensional quantum gravity are analyzed by enumerating rooted orientable maps. The Penner Model involves all smooth maps. φ − 4 model involves only 4-regular maps.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 17 / 38

slide-52
SLIDE 52

2-dimensional Quantum Gravity

Two models of 2-dimensional quantum gravity are analyzed by enumerating rooted orientable maps. The Penner Model involves all smooth maps. φ − 4 model involves only 4-regular maps.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 17 / 38

slide-53
SLIDE 53

2-dimensional Quantum Gravity

Two models of 2-dimensional quantum gravity are analyzed by enumerating rooted orientable maps. The Penner Model involves all smooth maps. φ − 4 model involves only 4-regular maps. The models have the same behaviour.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 17 / 38

slide-54
SLIDE 54

An algebraic explanation - A remarkable identity

Theorem (Jackson and Visentin)

Q(u2, x, y, z) = 1

2M(4u2, y + u, y, xz2) + 1 2M(4u2, y − u, y, xz2)

= biseven u M(4u2, y + u, y, xz2) M is the genus series for rooted orientable maps, and Q is the corresponding series for 4-regular maps. M(u2, x, y, z) :=

  • m∈M

u2g(m)xv(m)yf(m)ze(m) Q(u2, x, y, z) :=

  • m∈Q

u2g(m)xv(m)yf(m)ze(m) g(m), v(m), f(m), and e(m) are genus, #vertices, #faces, and #edges

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 18 / 38

slide-55
SLIDE 55

An algebraic explanation - A remarkable identity

Theorem (Jackson and Visentin)

Q(u2, x, y, z) = 1

2M(4u2, y + u, y, xz2) + 1 2M(4u2, y − u, y, xz2)

= biseven u M(4u2, y + u, y, xz2) The right hand side is a generating series for a set ¯ M consisting of elements of M with each handle decorated independently in one of 4 ways, and an even subset of vertices marked.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 18 / 38

slide-56
SLIDE 56

An algebraic explanation - A remarkable identity

Theorem (Jackson and Visentin)

Q(u2, x, y, z) = 1

2M(4u2, y + u, y, xz2) + 1 2M(4u2, y − u, y, xz2)

q -Conjecture (Jackson and Visentin)

The identity is explained by a natural bijection φ from ¯ M to Q. A decorated map with v vertices 2k marked vertices e edges f faces genus g

φ

A 4-regular map with e vertices 2e edges f + v − 2k faces genus g + k

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 18 / 38

slide-57
SLIDE 57

Two Clues

The radial construction for undecorated maps One extra image of φ

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 19 / 38

slide-58
SLIDE 58

Two Clues

The radial construction for undecorated maps One extra image of φ

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 19 / 38

slide-59
SLIDE 59

Two Clues

One extra image of φ

, [ ] , [ ]

= =

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 19 / 38

slide-60
SLIDE 60

A refined q-Conjecture

Conjecture (La Croix)

There is a natural bijection φ from ¯ M to Q such that: A decorated map with v vertices 2k marked vertices e edges f faces genus g

φ

A 4-regular map with e vertices 2e edges f + v − 2k faces genus g + k and the root edge of φ(m) is face-separating if and only if the root vertex of m is not decorated.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 20 / 38

slide-61
SLIDE 61

Root vertices in ¯ M are related to root edges in Q

Example (planar maps with 2 edges and 2 decorated vertices)

Nine of eleven rooted maps have a decorated root vertex.

Example (4-regular maps on the torus with two vertices)

Nine of fifteen rooted maps have face-non-separating root edges.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 21 / 38

slide-62
SLIDE 62

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 21 / 38

slide-63
SLIDE 63

The Map Series

Extra symmetry makes it easier to work with a more refined series. An enumerative problem associated with maps is to determine the number

  • f rooted maps with specified vertex- and face- degree partitions.

Definition

The map series for a set M of rooted maps is the combinatorial sum M(x, y, z) :=

  • m∈M

xν(m)yφ(m)z|E(m)| where ν(m) and φ(m) are the the vertex- and face-degree partitions of m.

Example

Rootings of are enumerated by

  • x3

2

x3

2

x3

2

x3

2

x3

2 x2 3

x2

3

x2

3

x2

3

x2

3

  • (y3

y3 y3 y3 y3 y4 y4 y4 y4 y4 y5 y5 y5 y5 y5) z6.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 22 / 38

slide-64
SLIDE 64

The Map Series

Extra symmetry makes it easier to work with a more refined series. An enumerative problem associated with maps is to determine the number

  • f rooted maps with specified vertex- and face- degree partitions.

Definition

The map series for a set M of rooted maps is the combinatorial sum M(x, y, z) :=

  • m∈M

xν(m)yφ(m)z|E(m)| where ν(m) and φ(m) are the the vertex- and face-degree partitions of m.

Example

Rootings of are enumerated by

  • x3

2

x3

2

x3

2

x3

2

x3

2 x2 3

x2

3

x2

3

x2

3

x2

3

  • (y3

y3 y3 y3 y3 y4 y4 y4 y4 y4 y5 y5 y5 y5 y5) z6.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 22 / 38

slide-65
SLIDE 65

Encoding Orientable Maps

1 Orient and label the edges. 2 This induces labels on flags. 3 Clockwise circulations at each

vertex determine ν.

4 Face circulations are the cycles

  • f ǫν.

1 2 3 4 5 6

ǫ = (1 1′)(2 2′)(3 3′)(4 4′)(5 5′)(6 6′) ν = (1 2 3)(1′ 4)(2′ 5)(3′ 5′ 6)(4′ 6′) ǫν = φ = (1 4 6′ 3′)(1′ 2 5 6 4′)(2′ 3 5′)

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 23 / 38

slide-66
SLIDE 66

Encoding Orientable Maps

1 Orient and label the edges. 2 This induces labels on flags. 3 Clockwise circulations at each

vertex determine ν.

4 Face circulations are the cycles

  • f ǫν.

1 2 3 4 5 6

ǫ = (1 1′)(2 2′)(3 3′)(4 4′)(5 5′)(6 6′) ν = (1 2 3)(1′ 4)(2′ 5)(3′ 5′ 6)(4′ 6′) ǫν = φ = (1 4 6′ 3′)(1′ 2 5 6 4′)(2′ 3 5′)

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 23 / 38

slide-67
SLIDE 67

Encoding Orientable Maps

1 Orient and label the edges. 2 This induces labels on flags. 3 Clockwise circulations at each

vertex determine ν.

4 Face circulations are the cycles

  • f ǫν.

1’ 1 2’ 2 3’ 3 4’ 4 5’ 5 6’ 6

ǫ = (1 1′)(2 2′)(3 3′)(4 4′)(5 5′)(6 6′) ν = (1 2 3)(1′ 4)(2′ 5)(3′ 5′ 6)(4′ 6′) ǫν = φ = (1 4 6′ 3′)(1′ 2 5 6 4′)(2′ 3 5′)

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 23 / 38

slide-68
SLIDE 68

Encoding Orientable Maps

1 Orient and label the edges. 2 This induces labels on flags. 3 Clockwise circulations at each

vertex determine ν.

4 Face circulations are the cycles

  • f ǫν.

1’ 1 2’ 2 3’ 3 4’ 4 5’ 5 6’ 6

ǫ = (1 1′)(2 2′)(3 3′)(4 4′)(5 5′)(6 6′) ν = (1 2 3)(1′ 4)(2′ 5)(3′ 5′ 6)(4′ 6′) ǫν = φ = (1 4 6′ 3′)(1′ 2 5 6 4′)(2′ 3 5′)

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 23 / 38

slide-69
SLIDE 69

A M¨

  • bius Strip

Maps can also be drawn in surfaces that contain M¨

  • bius strips.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 24 / 38

slide-70
SLIDE 70

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Start with a ribbon graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-71
SLIDE 71

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Start with a ribbon graph.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-72
SLIDE 72

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Mv Me Mf Ribbon boundaries determine 3 perfect matchings of flags.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-73
SLIDE 73

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Mv Me Pairs of matchings determine, faces,

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-74
SLIDE 74

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Mv Mf Pairs of matchings determine, faces, edges,

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-75
SLIDE 75

Encoding Locally Orientable Maps

A new encoding is needed to record twisting. Me Mf Pairs of matchings determine, faces, edges, and vertices.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-76
SLIDE 76

Encoding Locally Orientable Maps

1 1’ 2 2’ 3 3’ 4 4’ 5 5’ 6 6’ 7 7’ 8 8’

Mv Me Mf Mv =

  • {1, 3}, {1′, 3′}, {2, 5}, {2′, 5′}, {4, 8′}, {4′, 8}, {6, 7}, {6′, 7′}
  • Me =
  • {1, 2′}, {1′, 4}, {2, 3′}, {3, 4′}, {5, 6′}, {5′, 8}, {6, 7′}, {7, 8′}
  • Mf =
  • {1, 1′}, {2, 2′}, {3, 3′}, {4, 4′}, {5, 5′}, {6, 6′}, {7, 7′}, {8, 8′}
  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 25 / 38

slide-77
SLIDE 77

Hypermaps

Generalizing the combinatorial encoding, an arbitrary triple of perfect matchings determines a hypermap when the triple induces a connected graph, with cycles of Me ∪ Mf, Me ∪ Mv, and Mv ∪ Mf determining vertices, hyperfaces, and hyperedges.

Example

Hypermaps both specialize and generalize maps.

Example

֒ →

Hypermaps can be represented as face-bipartite maps.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 26 / 38

slide-78
SLIDE 78

Hypermaps

Generalizing the combinatorial encoding, an arbitrary triple of perfect matchings determines a hypermap when the triple induces a connected graph, with cycles of Me ∪ Mf, Me ∪ Mv, and Mv ∪ Mf determining vertices, hyperfaces, and hyperedges.

Example

Hypermaps both specialize and generalize maps.

Example

֒ →

Maps can be represented as hypermaps with ǫ = [2n].

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 26 / 38

slide-79
SLIDE 79

The Hypermap Series

Definition

The hypermap series for a set H of hypermaps is the combinatorial sum H(x, y, z) :=

  • h∈H

xν(h)yφ(h)zǫ(h) where ν(h), φ(h), and ǫ(h) are the vertex-, hyperface-, and hyperedge- degree partitions of h.

Example

Note

M(x, y, z) = H(x, y, z)

  • zi=zδi,2

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 27 / 38

slide-80
SLIDE 80

The Hypermap Series

Definition

The hypermap series for a set H of hypermaps is the combinatorial sum H(x, y, z) :=

  • h∈H

xν(h)yφ(h)zǫ(h) where ν(h), φ(h), and ǫ(h) are the vertex-, hyperface-, and hyperedge- degree partitions of h.

Example

Note

M(x, y, z) = H(x, y, z)

  • zi=zδi,2

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 27 / 38

slide-81
SLIDE 81

How does this help?

Instead of counting rooted maps, we can count labelled hypermaps. The numbers are different, but the correction factor is easy. Labelled counting problems are turned into problems involving counting factorizations. These can be answered via character theory. Appropriate characters appear as coefficients of symmetric functions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 28 / 38

slide-82
SLIDE 82

How does this help?

Instead of counting rooted maps, we can count labelled hypermaps. The numbers are different, but the correction factor is easy. Labelled counting problems are turned into problems involving counting factorizations. These can be answered via character theory. Appropriate characters appear as coefficients of symmetric functions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 28 / 38

slide-83
SLIDE 83

How does this help?

Instead of counting rooted maps, we can count labelled hypermaps. The numbers are different, but the correction factor is easy. Labelled counting problems are turned into problems involving counting factorizations. These can be answered via character theory. Appropriate characters appear as coefficients of symmetric functions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 28 / 38

slide-84
SLIDE 84

How does this help?

Instead of counting rooted maps, we can count labelled hypermaps. The numbers are different, but the correction factor is easy. Labelled counting problems are turned into problems involving counting factorizations. These can be answered via character theory. Appropriate characters appear as coefficients of symmetric functions.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 28 / 38

slide-85
SLIDE 85

Explicit Formulae

The hypermap series can be computed explicitly when H consists of all

  • rientable or locally orientable hypermaps.

Theorem (Jackson and Visentin)

When H is the set of orientable hypermaps, H

  • p(x), p(y), p(z); 0
  • = t ∂

∂t ln

  • θ∈P

Hθsθ(x)sθ(y)sθ(z)

  • t=0.

Theorem (Goulden and Jackson)

When H is the set of locally orientable hypermaps, H

  • p(x), p(y), p(z); 1
  • = 2t ∂

∂t ln

  • θ∈P

1 H2θ Zθ(x)Zθ(y)Zθ(z)

  • t=0.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 29 / 38

slide-86
SLIDE 86

Outline

1

Combinatorial Enumeration

2

Graphs, Maps, and Surfaces

3

Rooted Maps and Flags

4

Quantum gravity and the q-Conjecture

5

Map Enumeration Orientable Maps Non-Orientable Maps Hypermaps Generating Series

6

What does Jack have to do with it? The invariants resolve a special case

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 29 / 38

slide-87
SLIDE 87

Parallel Problems

We started with two similar problems, applied similar techniques, and found similar looking solutions. The natural question is, “Could we have solved both problems at once?”

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 30 / 38

slide-88
SLIDE 88

Jack Symmetric Functions

Jack symmetric functions,

Definition , are a one-parameter family, denoted

by {Jθ(α)}θ, that generalizes both Schur functions and zonal polynomials.

Proposition (Stanley)

Jack symmetric functions are related to Schur functions and zonal polynomials by: Jλ(1) = Hλsλ, Jλ, Jλ1 = H2

λ,

Jλ(2) = Zλ, and Jλ, Jλ2 = H2λ, where 2λ is the partition obtained from λ by multiplying each part by two.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 31 / 38

slide-89
SLIDE 89

A Generalized Series

b -Conjecture (Goulden and Jackson)

The generalized series, H

  • p(x), p(y), p(z); b
  • := (1 + b)t ∂

∂t ln

  • θ∈P

Jθ(x; 1 + b)Jθ(y; 1 + b)Jθ(z; 1 + b) Jθ, Jθ1+b

  • t=0

=

  • n≥0
  • ν,φ,ǫ⊢n

cν,φ,ǫ(b)pν(x)pφ(y)pǫ(z), has an combinatorial interpretation involving hypermaps. In particular cν,φ,ǫ(b) =

  • h∈Hν,φ,ǫ

b β(h) for some invariant β of rooted hypermaps.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 32 / 38

slide-90
SLIDE 90

b is ubiquitous

The many lives of b

b = 0 b = 1 Hypermaps Orientable ? Locally Orientable Symmetric Functions sθ Jθ(b) Zθ Matrix Integrals Hermitian ? Real Symmetric Moduli Spaces

  • ver C

?

  • ver R

Matching Systems Bipartite ? All

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 33 / 38

slide-91
SLIDE 91

A b-Invariant

The b-Conjecture assumes that cν,φ,ǫ(b) is a polynomial, and numerical evidence suggests that its degree is the genus of the hypermaps it

  • enumerates. A b-invariant must:

1 be zero for orientable hypermaps, 2 be positive for non-orientable hypermaps, and 3 depend on rooting.

Example

Rootings

  • f

precisely three maps are enumerated by c[4],[4],[22](b) = 1 + b + 3b2.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 34 / 38

slide-92
SLIDE 92

A b-Invariant

The b-Conjecture assumes that cν,φ,ǫ(b) is a polynomial, and numerical evidence suggests that its degree is the genus of the hypermaps it

  • enumerates. A b-invariant must:

1 be zero for orientable hypermaps, 2 be positive for non-orientable hypermaps, and 3 depend on rooting.

Example

Rootings

  • f

precisely three maps are enumerated by c[4],[4],[22](b) = 1 + b + 3b2.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 34 / 38

slide-93
SLIDE 93

A b-Invariant

The b-Conjecture assumes that cν,φ,ǫ(b) is a polynomial, and numerical evidence suggests that its degree is the genus of the hypermaps it

  • enumerates. A b-invariant must:

1 be zero for orientable hypermaps, 2 be positive for non-orientable hypermaps, and 3 depend on rooting.

Example

Rootings

  • f

precisely three maps are enumerated by c[4],[4],[22](b) = 1 + b + 3b2.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 34 / 38

slide-94
SLIDE 94

A b-Invariant

The b-Conjecture assumes that cν,φ,ǫ(b) is a polynomial, and numerical evidence suggests that its degree is the genus of the hypermaps it

  • enumerates. A b-invariant must:

1 be zero for orientable hypermaps, 2 be positive for non-orientable hypermaps, and 3 depend on rooting.

Example

Rootings

  • f

precisely three maps are enumerated by c[4],[4],[22](b) = 1 + b + 3b2.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 34 / 38

slide-95
SLIDE 95

A b-Invariant

The b-Conjecture assumes that cν,φ,ǫ(b) is a polynomial, and numerical evidence suggests that its degree is the genus of the hypermaps it

  • enumerates. A b-invariant must:

1 be zero for orientable hypermaps, 2 be positive for non-orientable hypermaps, and 3 depend on rooting.

Example

Rootings

  • f

precisely three maps are enumerated by c[4],[4],[22](b) = 1 + b + 3b2.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 34 / 38

slide-96
SLIDE 96

A root-edge classification

There are four possible types of root edges in a map.

Borders Bridges Handles Cross-Borders Example

A handle

Example

A cross-border

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 35 / 38

slide-97
SLIDE 97

A root-edge classification

There are four possible types of root edges in a map.

Borders Bridges Handles Cross-Borders Example

A handle

Example

A cross-border

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 35 / 38

slide-98
SLIDE 98

A root-edge classification

Handles occur in pairs

e e’

Untwisted Twisted

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 35 / 38

slide-99
SLIDE 99

A family of invariants

The invariant η

Iteratively deleting the root edge assigns a type to each edge in a map. An invariant, η, is given by η(m) := (# of cross-borders) + (# of twisted handles) . Different handle twisting determines a different invariant.

Example

Handle Cross-Border Border

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 36 / 38

slide-100
SLIDE 100

A family of invariants

The invariant η

Iteratively deleting the root edge assigns a type to each edge in a map. An invariant, η, is given by η(m) := (# of cross-borders) + (# of twisted handles) . Different handle twisting determines a different invariant.

Example

Handle Cross-Border Border

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 36 / 38

slide-101
SLIDE 101

A family of invariants

The invariant η

Iteratively deleting the root edge assigns a type to each edge in a map. An invariant, η, is given by η(m) := (# of cross-borders) + (# of twisted handles) . Different handle twisting determines a different invariant.

Example

Handle Cross-Border Border

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 36 / 38

slide-102
SLIDE 102

Main result (marginal b-invariants exist)

Theorem (La Croix)

If φ partitions 2n and η is a member of the family of invariants then, dv,φ(b) :=

  • ℓ(ν)=v

cν,φ,[2n](b) =

  • m∈Mv,φ

bη(m).

Proof (sketch).

Distinguish between root and non-root faces in the generating series. Show that this series satisfies a PDE with a unique solution. Predict an expression for the corresponding algebraic refinement. Show that the refined series satisfies the same PDE.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 37 / 38

slide-103
SLIDE 103

Main result (marginal b-invariants exist)

Theorem (La Croix)

If φ partitions 2n and η is a member of the family of invariants then, dv,φ(b) :=

  • ℓ(ν)=v

cν,φ,[2n](b) =

  • m∈Mv,φ

bη(m).

Proof (sketch).

Distinguish between root and non-root faces in the generating series. Show that this series satisfies a PDE with a unique solution. Predict an expression for the corresponding algebraic refinement. Show that the refined series satisfies the same PDE.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 37 / 38

slide-104
SLIDE 104

Main result (marginal b-invariants exist)

Theorem (La Croix)

If φ partitions 2n and η is a member of the family of invariants then, dv,φ(b) :=

  • ℓ(ν)=v

cν,φ,[2n](b) =

  • m∈Mv,φ

bη(m).

Proof (sketch).

Distinguish between root and non-root faces in the generating series. Show that this series satisfies a PDE with a unique solution. Predict an expression for the corresponding algebraic refinement. Show that the refined series satisfies the same PDE.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 37 / 38

slide-105
SLIDE 105

Main result (marginal b-invariants exist)

Theorem (La Croix)

If φ partitions 2n and η is a member of the family of invariants then, dv,φ(b) :=

  • ℓ(ν)=v

cν,φ,[2n](b) =

  • m∈Mv,φ

bη(m).

Proof (sketch).

Distinguish between root and non-root faces in the generating series. Show that this series satisfies a PDE with a unique solution. Predict an expression for the corresponding algebraic refinement. Show that the refined series satisfies the same PDE.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 37 / 38

slide-106
SLIDE 106

Main result (marginal b-invariants exist)

Theorem (La Croix)

If φ partitions 2n and η is a member of the family of invariants then, dv,φ(b) :=

  • ℓ(ν)=v

cν,φ,[2n](b) =

  • m∈Mv,φ

bη(m).

Proof (sketch).

Distinguish between root and non-root faces in the generating series. Show that this series satisfies a PDE with a unique solution. Predict an expression for the corresponding algebraic refinement. Show that the refined series satisfies the same PDE.

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 37 / 38

slide-107
SLIDE 107

Finding a partial differential equation

Root-edge type Schematic Contribution to M Cross-border z

  • i≥0

(i + 1)bri+2 ∂ ∂ri M Border z

  • i≥0

i+1

  • j=1

rjyi−j+2 ∂ ∂ri M Handle z

  • i,j≥0

(1 + b)jri+j+2 ∂2 ∂ri∂yj M Bridge z

  • i,j≥0

ri+j+2 ∂ ∂ri M ∂ ∂rj M

  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-108
SLIDE 108

Finding a partial differential equation

Root-edge type Schematic Contribution to M Cross-border z

  • i≥0

(i + 1)bri+2 ∂ ∂ri M Border z

  • i≥0

i+1

  • j=1

rjyi−j+2 ∂ ∂ri M Handle z

  • i,j≥0

(1 + b)jri+j+2 ∂2 ∂ri∂yj M Bridge z

  • i,j≥0

ri+j+2 ∂ ∂ri M ∂ ∂rj M

  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-109
SLIDE 109

Finding a partial differential equation

Root-edge type Schematic Contribution to M Cross-border z

  • i≥0

(i + 1)bri+2 ∂ ∂ri M Border z

  • i≥0

i+1

  • j=1

rjyi−j+2 ∂ ∂ri M Handle z

  • i,j≥0

(1 + b)jri+j+2 ∂2 ∂ri∂yj M Bridge z

  • i,j≥0

ri+j+2 ∂ ∂ri M ∂ ∂rj M

  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-110
SLIDE 110

Finding a partial differential equation

Root-edge type Schematic Contribution to M Cross-border z

  • i≥0

(i + 1)bri+2 ∂ ∂ri M Border z

  • i≥0

i+1

  • j=1

rjyi−j+2 ∂ ∂ri M Handle z

  • i,j≥0

(1 + b)jri+j+2 ∂2 ∂ri∂yj M Bridge z

  • i,j≥0

ri+j+2 ∂ ∂ri M ∂ ∂rj M

  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-111
SLIDE 111

Finding a partial differential equation

Root-edge type Schematic Contribution to M Cross-border z

  • i≥0

(i + 1)bri+2 ∂ ∂ri M Border z

  • i≥0

i+1

  • j=1

rjyi−j+2 ∂ ∂ri M Handle z

  • i,j≥0

(1 + b)jri+j+2 ∂2 ∂ri∂yj M Bridge z

  • i,j≥0

ri+j+2 ∂ ∂ri M ∂ ∂rj M

  • Michael La Croix (University of Waterloo)

The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-112
SLIDE 112

The End

Thank You

  • 5
  • 4
  • 3
  • 2
  • 1

1 2 3 4

x-axis

  • 4
  • 3
  • 2
  • 1

1 2 3

y-axis

1 2 3 4 5 6

z-axis

Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 38 / 38

slide-113
SLIDE 113

Example

Mf Mv Me ν = [23] ǫ = [32] φ = [6]

Return Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 39 / 38

slide-114
SLIDE 114

Example

is enumerated by

  • x3

2 x2 3

  • (y3 y4 y5)
  • z6

2

  • .

ν = [23, 32] φ = [3, 4, 5] ǫ = [26]

Return Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 40 / 38

slide-115
SLIDE 115

Jack Symmetric Functions

With respect to the inner product defined by pλ(x), pµ(x) = δλ,µ |λ|! |Cλ|αℓ(λ), Jack symmetric functions are the unique family satisfying: (P1) (Orthogonality) If λ = µ, then Jλ, Jµα = 0. (P2) (Triangularity) Jλ =

µλ vλµ(α)mµ, where vλµ(α) is a rational

function in α, and ‘’ denotes the natural order on partitions. (P3) (Normalization) If |λ| = n, then vλ,[1n](α) = n!.

Return Michael La Croix (University of Waterloo) The Jack parameter and Map Enumeration February 15, 2011 41 / 38