Cops and Robbers 1
Cops and Robbers: Directions and Generalizations
Anthony BonatoRyerson University
GRASTA 2012
Cops and Robbers
• played on reflexive undirected graphs G• two players Cops C and robber R play at alternate
time-steps (cops first) with perfect information• players move to vertices along edges; allowed to
moved to neighbors or pass • cops try to capture (i.e. land on) the robber, while
robber tries to evade capture• minimum number of cops needed to capture the
robber is the cop number c(G)– well-defined as c(G) ≤ |V(G)|
Cops and Robbers 6
Cops and Robbers 7
Basic facts on the cop number
• c(G) ≤ γ(G) (the domination number of G)– far from sharp: paths
• trees have cop number 1– one cop chases the robber to an end-vertex
• cop number can vary drastically with subgraphs– add a universal vertex
How big can the cop number be?
• c(n) = maximum cop number of a connected
graph of order n
• Meyniel’s Conjecture: c(n) = O(n1/2).
Cops and Robbers 8
State-of-the-art
• (Lu, Peng, 12) proved that
– independently proved by (Scott, Sudakov,11) and
(Frieze, Krivelevich, Loh, 11)
• (Bollobás, Kun, Leader, 12+): if
p = p(n) ≥ 2.1log n/ n, then
c(G(n,p)) ≤ 160000n1/2log n
• (Prałat,Wormald,12+): removed log factor
Cops and Robbers 11
)1(1
log))1(1( 22)( o
non
nOnc
Graph classes
• (Aigner, Fromme,84): Planar (outerplanar) graphs have cop number at most 3 (2).
• (Andreae,86): H-minor free graphs have cop number bounded by a constant.
• (Joret et al,10): H-free class graphs have bounded cop number iff each component of H is a tree with at most 3 leaves.
• (Lu,Peng,12): Meyniel’s conjecture holds for diameter 2 graphs, bipartite diameter 3 graphs.
Cops and Robbers 12
Cops and Robbers 13
How close to n1/2?
• consider a finite projective plane P– two lines meet in a unique point– two points determine a unique line– exist 4 points, no line contains more than two of them
• q2+q+1 points; each line (point) contains (is incident with) q+1 points (lines)
• incidence graph (IG) of P:– bipartite graph G(P) with red nodes the points of P
and blue nodes the lines of P– a point is joined to a line if it is on that line
Meyniel extremal families
• a family of connected graphs (Gn: n ≥ 1) is Meyniel extremal (ME) if there is a constant d > 0, such that for all n ≥ 1, c(Gn) ≥ dn1/2
• IG of projective planes: girth 6, (q+1)-regular, so have cop number ≥ q+1– order 2(q2+q+1)– Meyniel extremal (must fill in non-prime orders)– (Prałat,10) cop number = q+1
Cops and Robbers 15
Diameter 2
• (Lu, Peng, 12): If G has diameter 2, then c(G) ≤ 2n1/2 - 1.
– diameter 2 graphs satisfy Meyniel’s conjecture
• proof uses the probabilistic method
• Question: are there explicit Meyniel extremal families whose members are diameter two?
Cops and Robbers 16
Polarity graphs• suppose PG(2,q) has points P and lines L. A polarity is a
function π: P→ L such that for all points p,q, p ϵ π(q) iff q ϵ π(p).
• eg of orthogonal polarity: point mapped to its orthogonal complement
• polarity graph: vertices are points, x and y adjacent if
xϵ π(y)
Cops and Robbers 17
Properties of polarity graphs
• order q2+q+1• (q,q+1)-regular• C4-free
• (Erdős, Rényi, Sós,66), (Brown,66) orthogonal polarity graphs C4-free extremal
• diameter 2• (Godsil, Newman, 2008) have unbounded
chromatic number as q→ ∞
Cops and Robbers 18
Meyniel Extremal
• Theorem (B,Burgess,12+) Let q be a prime power. If Gq is a polarity graph of PG(2, q), then
q/2 ≤ c(Gq) ≤ q + 1.
• lower bound: lemma• upper bound: direct analysis
Cops and Robbers 19
Lower bounds
Lemma (Aigner,Fromme, 1984)
If G is a connected graph of girth at least 5, then c(G) ≥ δ(G).
Lemma (BB,12+)
If G is connected and K2,t-free, then
c(G) ≥ δ(G) / t.– applies to polarity graphs: t = 2
Cops and Robbers 21
t-orbit graphs
• (Füredi,1996) � described a family of K2,t+1-free extremal graphs of order (q2 -1)/t and which are (q,q+1)-regular for prime powers q.
• gives rise to a new family of ME graphs which are K2,t+1-free
Cops and Robbers 25
BIBDs
• a BIBD(v, k, λ) is a pair (V, B), where V is a set of v points, and B is a set of k-subsets of V, called blocks, such that each pair of points is contained in exactly λ blocks.
• Theorem (BB,12+) The cop number of the IG of a BIBD(v, k, λ) is between k and r, the replication number.
Cops and Robbers 27
Sketch of proof
• lower bound: girth 6; apply AF lemma and Fisher’s inequality
• upper bound:
Cops and Robbers 28
C
R
Block Intersection graphs
• given a block design (V,B), its block intersection graph has vertices equalling blocks, with blocks adjacent if they intersect
Cops and Robbers 29
BIG cop number
Theorem (BB,12+) If G is the block intersection graph of a BIBD(v, k, 1), then c(G) ≤ k. If v > k(k-1)2 + 1, then c(G) = k.
• gives families with unbounded cop number; not ME
• also considered point graphs
Cops and Robbers 30
Questions
• Soft Meyniel’s conjecture: for some ε > 0,
c(n) = O(n1-ε).
• Meyniel’s conjecture in other graphs classes?– bounded chromatic number– bipartite graphs– diameter 3– claw-free
• ME families from something other than designs?– extremal graphs?
Cops and Robbers 31
R.J. Nowakowski, P. Winkler Vertex-to-vertex pursuit in a graph, Discrete Mathematics 43 (1983) 235-239.
• 5 pages• > 200 citations (most for either author)
Cops and Robbers 33
The NW relation
• (Nowakowski,Winkler,83) introduced a sequence of relations characterizing cop-win graphs
• u ≤0 v if u = v
• u ≤i v if for all x in N[u], there is a y in N[v] such that x ≤j y for some j < i.
Cops and Robbers 34
Characterization
• the relations are ≤i monotone increasing; thus, there is an integer k such that
≤k = ≤k+1 – write:
≤k = ≤
Theorem (Nowakowski, Winkler, 83)
A cop has a winning strategy iff ≤ is
V(G) x V(G).Cops and Robbers 36
k cops
• may define an analogous relation but in V(G) x V(Gk) (categorical product)
• (Clarke,MacGillivray,12) k cops have a winning strategy iff the relation ≤ is V(G) x V(Gk).
Cops and Robbers 37
Axioms for pursuit games• a pursuit game G is a discrete-time process satisfying the following:
1. Two players, Left L and Right R.
2. Perfect-information.
3. There is a set of allowed positions PL for L; similarly for Right.
4. For each state of the game and each player, there is a non-empty set of allowed moves. Each allowed move leaves the position of the other player unchanged.
5. There is a set of allowed start positions I a subset of PL x PR.
6. The game begins with L choosing some position pL and R choosing qR such that (pL, qR) is in I.
7. After each side has chosen its initial position, the sides move alternately with L moving first. Each side, on its turn, must choose an allowed move from its current position.
8. There is a subset of final positions, F. Left wins if at any time, the current position belongs to F. Right wins the current position never belongs to F.
Cops and Robbers 38
Examples of pursuit games1. Cops and Robbers
– play on graphs, digraphs, orders, hypergraphs, etc.– play at different speeds, or on different edge sets
2. Cops and Robbers with traps
3. Distance k Cops and Robbers
4. Tandem-win Cops and Robbers
5. Restricted Chess
6. Helicopter Cops and Robbers
7. Maker-Breaker Games
8. Seepage
9. Scared Robber
Cops and Robbers 39
Relational characterization
• given a pursuit game G, we may define relations on PL x PR as follows:
• pL ≤0 qR if (pL, qR) in F.
• pL ≤i qR if Right is on qR and for every xR in PR such that if Right has an allowed move from (pL, qR) to (pL, xR), there exists yL in PL such that xR ≤j yL for some j < i and Left has an allowed move from (pL, xR) to (yL, xR).
• define ≤ analogously as beforeCops and Robbers 40
Characterization
Theorem (B, MacGillivray,12) Left has a winning strategy in the a pursuit game G if and only if there exists pL in PL, which is the first component of an ordered pair in I, such that for all qR in PR with (pL, qR) in I there exists wL in the set of allowed moves for Left from pL such that qR ≤ wL.
• gives rise to a min/max expression for the length of the game
Cops and Robbers 41
Length of game
• for an allowed start position (pL, qR), define
Corollary (BM,12+) If Left has a winning strategy in the a pursuit game G, then assuming optimal play, the length of the game is
where IL is the set positions for Left which are the first component of an ordered pair in I.
Cops and Robbers 42
),(maxmin),(
RLqpp
qpRLLL
II
CGT
(Berlekamp, Conway, Guy, 82) A combinatorial game satisfies:
1. There are two players, Left and Right.
2. There is perfect information.
3. There is a set of allowed positions in the game.
4. The rules of the game specify how the game begins and, for each player and each position, which moves to other positions are allowed.
5. The players alternate moves.
6. The game ends when a position is reached where no moves are possible for the player whose turn it is to move. In normal play the last player to move wins.
Cops and Robbers 43
Pursuit → CGT
Theorem (BM,12+)
1. Every pursuit game is a combinatorial game.
2. Not every combinatorial game is a pursuit game.
• uses characterization of (Smith, 66) via game digraphs • Nim is a counter-example for item (2)
Cops and Robbers 45
Position independence
• a pursuit game G is position independent if: if the game is not over, the set of available moves for a side does not depend on the position of the other side.
• examples: Cops and Robbers …• non-examples: Helicopter Cops and
Robbers, Maker Breaker, …
Cops and Robbers 46
State digraph
• G a position independent pursuit game• GL= (PL, ML) and GR= (PR, MR) are the
position digraphs of G
• SG = GL x GR state digraph of G– not all edges make sense– ignore these
Cops and Robbers 47
Relational characterization
Corollary (BM,12+) Let G be a position independent pursuit game. If GL is strongly connected and there exists X in PL such that S = X x PR, then Left has a winning strategy in G if and only if ≤ = V (DG) = PL x PR .
• generalizes results of NW and CM
Cops and Robbers 48
Algorithm
Theorem (BM,12+) Let G be a position independent pursuit game. Given the graphs GL and GR, if N+
GL(PL)
and N+GR
(PR) can be obtained in time O(f(|PL|)) and O(g(|
PR)|), respectively, then there is a
O(|PL||PR|f(|PL|)g(|PR|)).
algorithm to determine if Left has a winning strategy.
Cops and Robbers 49
Eg: Cops and Robbers
• gives an
O(n2k+2)
algorithm to determine if k cops have a winning strategy
• matches best known algorithm (Clarke, MacGillivray,12)
Cops and Robbers 50
Cop-win graphs
• node u is a corner if there is a v such that N[v] contains N[u]
• a cop-win ordering of G is an enumeration (v1,v2,…,vn) of V(G) such that for all i < n, there is a j > i such that
Theorem (Nowakowski, Winkler 83; Quilliot, 78)
A graph is cop-win if and only if it has a cop-win ordering.idea: cop-win graphs always have corners; retract corner
and play shadow strategy;
- graphs with cop-win orderings are cop-win by induction
Cops and Robbers 51
][},...,,{][ 1 jniii vNvvvvN
Vertex elimination ordering
• cop-win orderings generalize to pursuit games:
Idea: – order vertices of state digraph SG
– removable vertices are those whose out-neighbours are “dominated” by some by a vertex with higher index in the sequence
Corollary (BM,12+): Left has a winning strategy in the position independent pursuit game G if and only if SG admits a removable vertex ordering.
Cops and Robbers 53
Infinite case
• analogous characterization holds• k such that ≤k = ≤k+1 is an ordinal, κ
– CR-ordinal• relational characterization and vertex-
ordering hold (now a transfinite sequence)
• NB: κ can be infinite:– cannot think of κ as length of game
Cops and Robbers 54
Questions• c(G)≤k? k fixed:
– optimizing complexity for small k?• c(G)≤k? k not fixed:
– (Fomin, Golovach, Kratochvíl, Nisse, Suchan, 08): NP-hard
– Goldstein, Reingold Conjecture: EXPTIME-complete.– Conjecture: not in NP.– PSPACE-complete?
• infinite graphs: what are the CR-ordinals κ for cop-win graphs? – same question, but for more than one cop
Cops and Robbers 56
Cops and Robbers 57
Another direction:Minimum orders
• Mk = minimum order of a k-cop-win graph
• M1 = 1, M2 = 4
• M3 = 10
(Baird, B, et al, 12+)
Cops and Robbers 58
Questions
• M4 = ? (the (4,5)-cage?)
• are the Mk monotone increasing?– for example, can it happen that M344 < M343?
• mk = minimum order of a connected G such that c(G) ≥ k
• (Baird, B, et al, 12+) mk = Ω(k2) is equivalent to Meyniel’s conjecture.
• mk = Mk for all k ≥ 4?
Good guys vs bad guys games in graphs
59
slow medium fast helicopter
slow traps, tandem-win
medium robot vacuum Cops and Robbers edge searching eternal security
fast cleaning distance k Cops and Robbers
Cops and Robbers on disjoint edge sets
The Angel and Devil
helicopter seepage Helicopter Cops and Robbers, Marshals, The Angel and Devil,Firefighter
Hex
badgood
Cops and Robbers