Informed search algorithms. Chapter 4, Sections 1 2 1

Transcription

1 Informed search algorithms Chapter 4, Sections 1 2 Chapter 4, Sections 1 2 1

2 Outline Best-first search A search Heuristics Chapter 4, Sections 1 2 2

3 Review: Tree search function Tree-Search( problem, fringe) returns a solution, or failure fringe Insert(Make-Node(Initial-State[problem]), fringe) loop do if fringe is empty then return failure node Remove-Front(fringe) if Goal-Test[problem] applied to State(node) succeeds return node fringe InsertAll(Expand(node, problem), fringe) A strategy is defined by picking the order of node expansion Chapter 4, Sections 1 2 3

4 Best-first search Idea: use an evaluation function for each node estimate of desirability Expand most desirable unexpanded node Implementation: fringe is a queue sorted in decreasing order of desirability Special cases: greedy search A search Chapter 4, Sections 1 2 4

5 Romania with step costs in km Oradea 71 Neamt Zerind Arad Timisoara 111 Lugoj 70 Mehadia Dobreta Sibiu 99 Fagaras 80 Rimnicu Vilcea 97 Pitesti Bucharest 90 Craiova Giurgiu 87 Iasi Urziceni Vaslui Hirsova 86 Eforie Straight line distance to Bucharest Arad 366 Bucharest 0 Craiova 160 Dobreta 242 Eforie 161 Fagaras 178 Giurgiu 77 Hirsova 151 Iasi 226 Lugoj 244 Mehadia 241 Neamt 234 Oradea 380 Pitesti 98 Rimnicu Vilcea 193 Sibiu 253 Timisoara 329 Urziceni 80 Vaslui 199 Zerind 374 Chapter 4, Sections 1 2 5

6 Greedy search Evaluation function h(n) (heuristic) = estimate of cost from n to the closest goal E.g., h SLD (n) = straight-line distance from n to Bucharest Greedy search expands the node that appears to be closest to goal Chapter 4, Sections 1 2 6

7 Greedy search example Arad 366 Chapter 4, Sections 1 2 7

8 Greedy search example Arad Sibiu Timisoara Zerind Chapter 4, Sections 1 2 8

9 Greedy search example Arad Sibiu Timisoara Zerind Arad Fagaras Oradea Rimnicu Vilcea Chapter 4, Sections 1 2 9

10 Greedy search example Arad Sibiu Timisoara Zerind Arad Fagaras Oradea Rimnicu Vilcea Sibiu Bucharest Chapter 4, Sections

11 Properties of greedy search Complete?? Chapter 4, Sections

12 Properties of greedy search Complete?? No can get stuck in loops, e.g., with Oradea as goal, Iasi Neamt Iasi Neamt Complete in finite space with repeated-state checking Time?? Chapter 4, Sections

13 Properties of greedy search Complete?? No can get stuck in loops, e.g., Iasi Neamt Iasi Neamt Complete in finite space with repeated-state checking Time?? O(b m ), but a good heuristic can give dramatic improvement Space?? Chapter 4, Sections

14 Properties of greedy search Complete?? No can get stuck in loops, e.g., Iasi Neamt Iasi Neamt Complete in finite space with repeated-state checking Time?? O(b m ), but a good heuristic can give dramatic improvement Space?? O(b m ) keeps all nodes in memory Optimal?? Chapter 4, Sections

15 Properties of greedy search Complete?? No can get stuck in loops, e.g., Iasi Neamt Iasi Neamt Complete in finite space with repeated-state checking Time?? O(b m ), but a good heuristic can give dramatic improvement Space?? O(b m ) keeps all nodes in memory Optimal?? No Chapter 4, Sections

16 A search Idea: avoid expanding paths that are already expensive Evaluation function f(n) = g(n) + h(n) g(n) = cost so far to reach n h(n) = estimated cost to goal from n f(n) = estimated total cost of path through n to goal A search uses an admissible heuristic i.e., h(n) h (n) where h (n) is the true cost from n. (Also require h(n) 0, so h(g) = 0 for any goal G.) E.g., h SLD (n) never overestimates the actual road distance Theorem: A search is optimal Chapter 4, Sections

17 A search example Arad 366=0+366 Chapter 4, Sections

18 A search example Arad Sibiu 393= Timisoara Zerind 447= = Chapter 4, Sections

19 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = = = Chapter 4, Sections

20 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras 646= = Oradea 671= Rimnicu Vilcea Craiova Pitesti Sibiu 526= = = Chapter 4, Sections

21 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = Sibiu Bucharest Craiova Pitesti Sibiu 591= = = = = Chapter 4, Sections

22 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = Sibiu Bucharest Craiova Pitesti Sibiu 591= = = = Bucharest Craiova Rimnicu Vilcea 418= = = Chapter 4, Sections

23 Optimality of A (standard proof) Suppose some suboptimal goal G 2 has been generated and is in the queue. Let n be an unexpanded node on a shortest path to an optimal goal G 1. Start n G G 2 f(g 2 ) = g(g 2 ) since h(g 2 ) = 0 > g(g 1 ) since G 2 is suboptimal f(n) since h is admissible Since f(g 2 ) > f(n), A will never select G 2 for expansion Chapter 4, Sections

24 Properties of A Complete?? Chapter 4, Sections

25 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f(g) Time?? Chapter 4, Sections

26 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f(g) Time?? Exponential in [relative error in h length of soln.] Space?? Chapter 4, Sections

27 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f(g) Time?? Exponential in [relative error in h length of soln.] Space?? Keeps all nodes in memory Optimal?? Chapter 4, Sections

28 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f(g) Time?? Exponential in [relative error in h length of soln.] Space?? Keeps all nodes in memory Optimal?? Yes cannot expand f i+1 until f i is finished A expands all nodes with f(n) < C A expands some nodes with f(n) = C A expands no nodes with f(n) > C Chapter 4, Sections

29 n previo Next: Example Up: l3 Previous: Optimality of A* IDA* Series of Depth-First Searches Like Iterative Deepening Search, except use A* cost threshold instead of depth threshold Ensures optimal solution queueing-fn is enqueue-at-front if f(child) threshold Threshold is h(root) for first pass Next threshold is f(min_child), where min_child is cutoff child with minimum f value This conservative increase ensures cannot look past optimal cost solution

30 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

31 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

32 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

33 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

34 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

35 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

36 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

37 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

38 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

39 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

40 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

41 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

42 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

43 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

44 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

45 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

46 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

47 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

48 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

49 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

50 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

51 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

52 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

53 n previo Next: Eight Puzzle Example Up: l3 Previous: IDA* Example Click mouse to advance to next frame.

54 n previo Next: Eight Puzzle Example Up: l3 Previous: Example Eight Puzzle Example

55 n previo Next: Eight Puzzle Example Up: l3 Previous: Eight Puzzle Example Eight Puzzle Example

56 n previo Next: Analysis Up: l3 Previous: Eight Puzzle Example Eight Puzzle Example

57 n previo Next: RBFS Up: l3 Previous: Eight Puzzle Example Analysis Some redundant search, but small amount compared to work done on last iteration Dangerous if f values are very close If threshold = 21.1 and next value is 21.2, probably only include 1 new node each iteration Time: Space: SMA* search can be used to remember some nodes from one iteration to the next.

58 A heuristic is consistent if Proof of lemma: Consistency h(n) c(n, a, n ) + h(n ) If h is consistent, we have f(n ) = g(n ) + h(n ) = g(n) + c(n, a, n ) + h(n ) g(n) + h(n) = f(n) I.e., f(n) is nondecreasing along any path. n c(n,a,n ) n h(n ) h(n) G Chapter 4, Sections

59 E.g., for the 8-puzzle: Admissible heuristics h 1 (n) = number of misplaced tiles h 2 (n) = total Manhattan distance (i.e., no. of squares from desired location of each tile) Start State Goal State h 1 (S) =?? h 2 (S) =?? Chapter 4, Sections

60 E.g., for the 8-puzzle: Admissible heuristics h 1 (n) = number of misplaced tiles h 2 (n) = total Manhattan distance (i.e., no. of squares from desired location of each tile) Start State Goal State h 1 (S) =?? 6 h 2 (S) =?? = 14 Chapter 4, Sections

61 Dominance If h 2 (n) h 1 (n) for all n (both admissible) then h 2 dominates h 1 and is better for search Typical search costs: d = 14 IDS = 3,473,941 nodes A (h 1 ) = 539 nodes A (h 2 ) = 113 nodes d = 24 IDS 54,000,000,000 nodes A (h 1 ) = 39,135 nodes A (h 2 ) = 1,641 nodes Given any admissible heuristics h a, h b, h(n) = max(h a (n), h b (n)) is also admissible and dominates h a, h b Chapter 4, Sections

62 Relaxed problems Admissible heuristics can be derived from the exact solution cost of a relaxed version of the problem If the rules of the 8-puzzle are relaxed so that a tile can move anywhere, then h 1 (n) gives the shortest solution If the rules are relaxed so that a tile can move to any adjacent square, then h 2 (n) gives the shortest solution Key point: the optimal solution cost of a relaxed problem is no greater than the optimal solution cost of the real problem Chapter 4, Sections

63 Relaxed problems contd. Well-known example: travelling salesperson problem (TSP) Find the shortest tour visiting all cities exactly once Minimum spanning tree can be computed in O(n 2 ) and is a lower bound on the shortest (open) tour Chapter 4, Sections

64 Summary Heuristic functions estimate costs of shortest paths Good heuristics can dramatically reduce search cost Greedy best-first search expands lowest h incomplete and not always optimal A search expands lowest g + h complete and optimal also optimally efficient (up to tie-breaks, for forward search) Admissible heuristics can be derived from exact solution of relaxed problems Chapter 4, Sections