Constraint satisfaction

Problems
 Constraint satisfaction Problems
• Till now we have seen that problems can be solved by searching in the
space of states
• We evaluate a state by defining a domain specific heuristic but each state is
atomic or indivisible
• CS strategy allows us to solve many problems more efficiently but we use a
factored representation for each state
• A state is defined by a set of variables and each of which has a value
• A problem is solved when each variable has a value such that it satisfies all
the constraints on the variable
• A problem described this way is called Constraint satisfaction Problem.
 Constraint satisfaction Problems
• CSP algorithms take advantage of the structure of states
• The objective is to eliminate large portions of the search space all at
once by identifying variable/value combinations that violate the
constraints
 Defining CSP
• A constraint satisfaction problem (or CSP) is defined by a set of variables, XI, X2,.
. . , Xn, and a set of constraints, C1, . . , Cn and a set of domains D1, D2…Dn.
• Each Domain Di consists of a set of allowable values{v1,v2…vn} for variable Xi
• Each Constraint Ci consists of a pair(scope, rel), where scope is a tuple of
variables which participate in the constraint and rel is a relation
• Relation defines the values that those variables can take on
• If the variables X1 and X2 both have domain{a,b}, then the constraint saying two
variables must have different values can be written as: <(X1,X2),[(a,b),(b,a)]> or
<(X1,X2),X1=! X2>
• Each step of solution of problem is defined by an assignment of values to some
or all of the variables.
• A solution to a CSP is a complete assignment that satisfies all the constraints.
Example: Graph Coloring problem
• Suppose we are looking at a map of Australia
• We are given the task of coloring each region either red, green or blue in
such a way that no two regions have the same color.
• we define the variables as : WA, NT, Q, NSW, V, SA, and T.
• The domain of each variable is: {red, green, blue).
• The constraints require neighboring regions to have distinct colors, so, the
allowable combinations for WA and NT are the pairs {(red, green), (red,
blue), (green, red), (green, blue), (blue, red), (blue, green)) .
• The constraint can also be represented as WA # NT
• There can be many possible solutions, such as: { WA= red, NT = green, Q =
red, NSW = green, V= red, SA= blue, T= red ).
Cont…
• A CSP can be formulated as an incremental standard search problem as
follows:
• Initial state: the empty assignment {), in which all variables are
unassigned.
• Successor function: a value can be assigned to any unassigned variable,
provided that it does not conflict with previously assigned variables.
• Goal test: the current assignment is complete.
• Path cost: a constant cost (e.g., 1) for every step.
• Every solution must be a complete assignment and therefore appears at
depth n if there are n variables.
• The search tree extends only to depth n and due to this depth first search
algorithms are popular for CSPs.
Cont..
• The simplest kind of CSP involves variables that are discrete and have finite
domains.
• Map-coloring problems and 8-queens problem are of this kind
• They can also be viewed as a finite-domain CSP.
• In 8-queens problem each queen is placed in columns 1, . . . ,8 and each row
variable has the domain {1,2,3,4,5,6,7,8).
• If the maximum domain size of any variable in a CSP is d, then the number of
possible complete assignments is O(d n )-that is, exponential in the number of
variables.
• Finite-domain CSPs include Boolean CSPs, whose variables can be either true or
false.
• Boolean CSPs include as special cases some NP-complete problems, such as 3SAT.
 Cryptarithmetic puzzles
• One of the famous example of CSPs is cryptarithmetic puzzles.
• Each letter in a cryptarithmetic puzzle represents a different digit.
• TWO
• +TWO
• ------------------------------------------
• FOUR
• ----------------------------------------------- O +O = R +10.C10
• C10 + W +W = U + 10. C100
• C100 + T +T = O + 10. C1000
• C1000 = F

-
 CSP formalism
• The simplest kind of CSP involves variables that have discrete, finite
domains
• Map coloring, job scheduling and 8 queen problems are of this kind
• A discrete domain can be infinite such as set of integers or strings
• With infinite domains like this, it is not possible to enumerate all
combinations of values.
• Mostly, constraints like T1+d1<=T2… are stated
• There are special solution algorithms for linear constraints on integer
variables(we do not discuss here) but no algo exists for nonlinear
constraints