1 using search in problem solving part ii. 2 basic concepts basic concepts: initial state...

29
1 Using Search in Using Search in Problem Solving Problem Solving Part II Part II

Post on 19-Dec-2015

224 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

1

Using Search in Using Search in Problem SolvingProblem Solving

Part IIPart II

Page 2: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

2

Basic ConceptsBasic ConceptsBasic concepts:

• Initial state

• Goal/Target state

• Intermediate states

• Path from the initial to the target state• Operators/rules to get from one state to another• All states - search space

We search for a path / sequence of operators to go from the initial state to the goal state

Page 3: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

3

Search methodsSearch methods

Basic (uninformed):Basic (uninformed): breadth-firstbreadth-first depth-firstdepth-first

Heuristic (informed): Heuristic (informed): hill climbing, hill climbing, best-first, best-first, A*A*

Page 4: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

4

Heuristic SearchHeuristic Search

Heuristic search is used to reduce the search space.

Basic idea: explore only promising states/paths.

We need an evaluation function to estimate each state/path.

Page 5: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

5

Hill climbingHill climbing

Basic idea: always head towards a state which is better than the current one.

There is no exhaustive search, so no node list is maintained.

If a solution is found, it is found for a very short time and with minimum memory requirements. However it is not guaranteed that a solution will be found - the local maxima problem.

Page 6: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

6

Hill Climbing - AlgorithmHill Climbing - Algorithm Start with current-state = initial-state. Until current-state = goal-state OR there is no

change in current-state do:

Get the successors of the current state and use the evaluation function to assign a score to each successor.

If one of the successors has a better score than the current-state then set the new current-state to be the successor with the best score.

Page 7: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

7

Hill ClimbingHill Climbing

Node list is not maintainedNode list is not maintained No problems with loops as we move No problems with loops as we move

to a better nodeto a better node If a solution is found, it is found for a If a solution is found, it is found for a

very short time with minimal very short time with minimal memory requirementsmemory requirements

Finding a solution is not guaranteed – Finding a solution is not guaranteed – the the local maxima problemlocal maxima problem

Page 8: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

8

Best First SearchBest First Search The node with the best score is chosen to

be expanded. Works in breadth-first manner, keeps a

data structure (called agenda, based on priority queues) of all successors and their scores.

If a node that has been chosen does not lead to a solution, the next "best" node is chosen, so eventually the solution is found

Always finds a solution, not guaranteed to be the optimal one.

Page 9: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

9

Best First Search Best First Search AlgorithmAlgorithm

1. Start with agenda = [initial-state].

2. While agenda is not empty do

A. Pick the best node on agenda.

B. If it is the goal node then return with success. Otherwise find its

successors.

C. Assign the successor nodes a score using the evaluation function and add the scored nodes to the agenda

Page 10: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

10

Comparison with hill-Comparison with hill-climbingclimbing

Similarities: best-first always chooses the best node

Difference: best-first search keeps an agenda as in breadth-first search, and in case of a dead end it will backtrack, choosing the next-best node.

Page 11: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

11

Beam SearchBeam Search

Best-first methodBest-first method

Expands the best few paths at each Expands the best few paths at each levellevel

Has the memory advantages of the Has the memory advantages of the depth-first searchdepth-first search

Not exhaustive and may not find the Not exhaustive and may not find the best solution. Finding a solution is best solution. Finding a solution is not guaranteednot guaranteed

Page 12: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

12

The A* AlgorithmThe A* AlgorithmAn evaluation function that accounts for - the cost of the paths - the score of the nodes

F(Node) = g(Node) + h(Node)

g(Node) - the costs from the initial state to the current node

h(Node) - future costs, i.e. node score

Disadvantage of A* is the memory requirement - the algorithm keeps records for the entire tree.

Page 13: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

13

The A* AlgorithmThe A* Algorithm

A* always finds the best

solution, provided that

h(Node) does not

overestimate the future

costs.

Page 14: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

14

Example

107

8

8

4

4

6

10

92

4

A

B: 9D : 10

F : 6

H : 5

C: 8E: 7

K: 5

2

G: 0

J: 4 Start: A

Goal: G

Page 15: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

15

Search and Problem-Search and Problem-solvingsolving

Generate and TestGenerate and Test Means-end analysis in planningMeans-end analysis in planning

work out the path backwardswork out the path backwards use rules with preconditions and effectsuse rules with preconditions and effects the target goal is the effect of some operator/rulethe target goal is the effect of some operator/rule set the conditions as subgoals to be achievedset the conditions as subgoals to be achieved

Minimax algorithm in games,Minimax algorithm in games, Alpha-Beta pruningAlpha-Beta pruning

Constraint satisfactionConstraint satisfaction

Page 16: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

16

Planning TechniquesPlanning Techniques

Means-Ends analysisInitial state, Goal state, intermediate states.Operators to change the states.

Each operator:Preconditions

Effects

Go backwards : from the goal to the initial state.

Page 17: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

17

Basic AlgorithmBasic AlgorithmPut the goal state in a list of subgoals

While the list is not empty, doTake a subgoal.

Is there an operation that has as an effect this goal state?

If no - the problem is not solvable.If yes, put the operation in the plan.

Are the preconditions fulfilled?If yes - we have constructed the planIf no - Put the preconditions in the list

of subgoals

Page 18: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

18

More complex problemsMore complex problems

Goal state - a set of sub-states

Each sub-state is a goal.

To achieve the target state we have to achieve each goal in the Target state.

Page 19: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

19

Extensions of the basic Extensions of the basic methodmethod

Planning with goal protectionPlanning with goal protection Nonlinear planningNonlinear planning Hierarchical planningHierarchical planning Reactive planningReactive planning

Page 20: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

20

Game Playing SystemsGame Playing Systems

Special feature of games: there is an opponent that tries to destroy our plans.

The search tree (game tree) has to reflect the possible moves of both players

Page 21: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

21

Player1

Player2

Player1

Page 22: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

22

Minimax procedureMinimax procedure

Scoring the final nodes: Winning : 10Draw : 0Loosing : -10

Each player is trying to maximize their score and minimize the score of the other player.

The levels in the tree for player P1 are maximizing if it is P1's move, and minimizing if it is P2's move.

Page 23: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

23

Minimax algorithmMinimax algorithm • Build the search tree.

• Assign scores to the leaves• Assign scores to intermediate nodes:

If node is a leaf of the search tree - return score of that nodeElse:

If it is a maximizing node return maximum of scores of the successor nodes

Else if it is a minimizing node - return minimum of scores of the successor nodes

After that, choose nodes with highest score

Page 24: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

24

Alpha-Beta PruningAlpha-Beta Pruning

At maximizing level the score is at least the score of any successor.

At minimizing level the score is at most the score of any successor.

Page 25: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

25

A:10

B:10 C: 0

E:10 F:10 G: 0 H:10

I:10 J: 0 K:10 L:-10 M:-10 N:0 O:10 P:-10

Maximizing score

Minimizing score

P1's move

P2's move

P1's minimax tree

Page 26: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

26

Constraint satisfaction Constraint satisfaction searchsearch

State description: a set of variables.A set of constraints on the assigned values of variables.

Each variable may be assigned a value within its domain, and the value must not violate the constraints.

Initial state - the variables do not have valuesGoal state: Variables have values and the constraints are obeyed.

Page 27: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

27

Constraint satisfaction Constraint satisfaction searchsearch

Different types of constraints: Different types of constraints:

unaryunary - on one variable only - on one variable only

binarybinary - on two variables, etc. - on two variables, etc.

Page 28: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

28

ExampleExample

WEEK LOAN WEEK + LOAN + WEEK

LOAN WEEK ----------------- -----------------

DEBT MONTH

Page 29: 1 Using Search in Problem Solving Part II. 2 Basic Concepts Basic concepts: Initial state Goal/Target state Intermediate states Path from the initial

29

Example - continuedExample - continued

The problem is to assign a digit to each letter, with different letters being different digits, so that when the letters are replaced by the digits, the addition is done correctly.

Initially no letters have values.

The goal state would be an assignment of values so that the addition is correct.

Operators: assign a digit to a letter, so that no two different digits have same values, and the addition rules are not violated.