COSC343: Artificial Intelligence
|
|
- Madison Ramsey
- 5 years ago
- Views:
Transcription
1 COSC343: Artificial Intelligence Lecture 18: Informed search algorithms Alistair Knott Dept. of Computer Science, University of Otago Alistair Knott (Otago) COSC343 Lecture 18 1 / 1
2 In today s lecture Best-first search A search Heuristics Alistair Knott (Otago) COSC343 Lecture 18 2 / 1
3 Review: Uninformed search Here s the general tree-search algorithm, as given in Lecture 13: 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) Recall: The search strategy depends on the way nodes are added to the fringe. (I.e. on the definition of INSERTALL.) Alistair Knott (Otago) COSC343 Lecture 18 3 / 1
4 Informed search The search algorithms we ve looked at so far have all been uninformed, or brute-force: they systematically explore the state space without any knowledge about the particular problem being solved. This means they re extremely inefficient. For most interesting problems, the state space is so large that brute force methods are far too slow. Often, knowledge about the problem can allow us to be a bit more smart about how to add nodes to the fringe. The basic idea: put nodes which are likely to lead to a good solution towards the front. Alistair Knott (Otago) COSC343 Lecture 18 4 / 1
5 Evaluation functions and best-first search To be formal, we can define an evaluation function f (n) on nodes, which takes a node in the search tree and returns a number. By convention, a low f (n) means a good candidate for expansion. We can now define a general type of search called best-first search, where the frontier is kept ordered in increasing order of f (n). Alistair Knott (Otago) COSC343 Lecture 18 5 / 1
6 Heuristics and greedy search What do we mean by a node which is likely to lead to a solution? If we knew for certain that it would lead to a solution, we wouldn t need to do the search. The principles we use for ordering the fringe are often just rules of thumb, or rules that work most of the time. A heuristic function gives an estimate of the cheapest path cost from the current state to a goal state. It s typically called h(n). Alistair Knott (Otago) COSC343 Lecture 18 6 / 1
7 Heuristics and greedy search What do we mean by a node which is likely to lead to a solution? If we knew for certain that it would lead to a solution, we wouldn t need to do the search. The principles we use for ordering the fringe are often just rules of thumb, or rules that work most of the time. A heuristic function gives an estimate of the cheapest path cost from the current state to a goal state. It s typically called h(n). In greedy search, the evaluation function is defined completely by a heuristic function. I.e. f (n) = h(n). Alistair Knott (Otago) COSC343 Lecture 18 6 / 1
8 An example of a heuristic function What counts as a good estimate of the shortest cost to the goal state? The answer depends on the domain. Recall the Romania state space - we re trying to get from Arad to Bucharest. A useful heuristic function here is the straight-line distance from a given town to the goal town (Bucharest). Does this function always give the best choice for the next node to expand? Does it normally give the best choice? Alistair Knott (Otago) COSC343 Lecture 18 7 / 1
9 Romania, plus straight-line distances to Bucharest Oradea 71 Neamt 87 Zerind Iasi Arad Sibiu 99 Fagaras Vaslui Timisoara Rimnicu Vilcea Lugoj 97 Pitesti Hirsova Mehadia Urziceni Bucharest 120 Dobreta 90 Craiova Giurgiu 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 Alistair Knott (Otago) COSC343 Lecture 18 8 / 1
10 Greedy search example Arad 366 Alistair Knott (Otago) COSC343 Lecture 18 9 / 1
11 Greedy search example Arad Sibiu Timisoara Zerind Alistair Knott (Otago) COSC343 Lecture 18 9 / 1
12 Greedy search example Arad Sibiu Timisoara Zerind Arad Fagaras Oradea Rimnicu Vilcea Alistair Knott (Otago) COSC343 Lecture 18 9 / 1
13 Greedy search example Arad Sibiu Timisoara Zerind Arad Fagaras Oradea Rimnicu Vilcea Sibiu Bucharest Alistair Knott (Otago) COSC343 Lecture 18 9 / 1
14 Properties of greedy search Complete?? Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
15 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest Complete in finite space with repeated-state checking Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
16 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest Complete in finite space with repeated-state checking Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
17 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest Complete in finite space with repeated-state checking Time?? O(b m ), but a good heuristic can give dramatic improvement Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
18 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest Complete in finite space with repeated-state checking Time?? O(b m ), but a good heuristic can give dramatic improvement Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
19 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest 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?? Alistair Knott (Otago) COSC343 Lecture / 1
20 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest 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?? Alistair Knott (Otago) COSC343 Lecture / 1
21 Properties of greedy search Complete?? No can get stuck in loops A B Bucharest 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 Alistair Knott (Otago) COSC343 Lecture / 1
22 A search The main problem with greedy search is that it only looks forward, to the goal state. To produce an optimal search, it s also necessary to keep track of the path cost from the start node to the current node. We can define a function called g(n), which returns the cost of the path from the start node to the current node. Note: g(n) is not an estimate: we know this bit! In A search, the evaluation function f (n) is defined as g(n) + h(n). A search is the starting point for all serious search algorithms. Alistair Knott (Otago) COSC343 Lecture / 1
23 A search example Oradea 71 Neamt 87 Zerind Iasi Arad Sibiu 99 Fagaras Vaslui Timisoara Rimnicu Vilcea Lugoj 97 Pitesti Hirsova Mehadia Urziceni Bucharest 120 Dobreta 90 Craiova Giurgiu 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 Alistair Knott (Otago) COSC343 Lecture / 1
24 A search example Arad 366=0+366 Alistair Knott (Otago) COSC343 Lecture / 1
25 A search example Arad Sibiu Timisoara Zerind 393= = = Alistair Knott (Otago) COSC343 Lecture / 1
26 A search example Arad Sibiu Timisoara Zerind Arad Fagaras Oradea Rimnicu Vilcea 646= = = = = = Alistair Knott (Otago) COSC343 Lecture / 1
27 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = = Craiova Pitesti Sibiu 526= = = Alistair Knott (Otago) COSC343 Lecture / 1
28 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = Sibiu Bucharest Craiova Pitesti Sibiu 591= = = = = Alistair Knott (Otago) COSC343 Lecture / 1
29 A search example Arad Sibiu Timisoara Zerind 447= = Arad Fagaras Oradea Rimnicu Vilcea 646= = Sibiu Bucharest Craiova Pietesti Sibiu 591= = = = Bucharest Craiova Rimnicu Vilcea 418= = = Alistair Knott (Otago) COSC343 Lecture / 1
30 Admissible heuristics One particularly interesting class of heuristics are those which are termed admissible. An admissible heuristic never over-estimates the cost of the path from the current state to the goal state. In other words, an admissible heuristic is always optimistic. Note that straight-line distance is an admissible heuristic. (Because a straight line is the shortest distance between two points.) Alistair Knott (Otago) COSC343 Lecture / 1
31 Optimality of A If A search uses an admissible heuristic, it is provably optimal: i.e. it is guaranteed to find the shortest path to the solution (if one exists). Here s the proof. Say G 1 is the optimal goal state, and G 2 is a suboptimal goal state which is already on the fringe. And say that n is another node on the fringe which is on the shortest path to G 1. We must guarantee that A will never pick G 2 from the fringe before picking n. We know that since G 2 is suboptimal, g(g 2 ) + h(g 2 ) is greater than the cost of the actual shortest path to a solution. Since h is admissible, we know that g(n) + h(n) is less than the actual shortest path to a solution. So A will never pick G 2 before n. Alistair Knott (Otago) COSC343 Lecture / 1
32 Graph search with A Recall that in tree search, we don t check for repeated states, while in graph search we only add a new node to the fringe if its state has not already been encountered. If we re interested in optimality, we must be careful when throwing away nodes in this way. - The two paths to the repeated state might have different costs. - We want to throw away the node with the higher g(n). We can do this by explicitly comparing the g(n) of each node. But another way is to ensure that the first time we reach a node, we always pick the shortest route. - We can prove this is the case if the heuristic used is consistent. Alistair Knott (Otago) COSC343 Lecture / 1
33 Consistent heuristics It s useful to have a heuristic that evaluates nodes along a path in a way that s consistent with path costs. Say a parent node has an h of 5, and it costs 3 to get to the child: child cost=3 parent h=? h=5 goal Alistair Knott (Otago) COSC343 Lecture / 1
34 Consistent heuristics It s useful to have a heuristic that evaluates nodes along a path in a way that s consistent with path costs. Say a parent node has an h of 5, and it costs 3 to get to the child: child cost=3 parent h=? h=5 goal It would be odd if the h value of the child was less than 5-3. It would be odd if the h value of the child was more than 5+3. Alistair Knott (Otago) COSC343 Lecture / 1
35 Consistent heuristics A consistent heuristic is one which guarantees for any parent p and child c that h(c) >= h(p) cost of path from p to c Alistair Knott (Otago) COSC343 Lecture / 1
36 Consistent heuristics A consistent heuristic is one which guarantees for any parent p and child c that h(c) >= h(p) cost of path from p to c Now think what that means in relation to the f scores of the parent and child. f (n) = g(n) + h(n)... g= h>=2 f>= child cost=3 h>=2 parent h=5 g=10 h=5 f= goal Alistair Knott (Otago) COSC343 Lecture / 1
37 Consistent heuristics A consistent heuristic is one which guarantees for any parent p and child c that h(c) >= h(p) cost of path from p to c Now think what that means in relation to the f scores of the parent and child. f (n) = g(n) + h(n)... g=10+3 h>=2 f>=15 child cost=3 h>=2 parent h=5 g=10 h=5 f=15 goal Alistair Knott (Otago) COSC343 Lecture / 1
38 Consistent heuristics A consistent heuristic is one which guarantees for any parent p and child c that h(c) >= h(p) cost of path from p to c Now think what that means in relation to the f scores of the parent and child. f (n) = g(n) + h(n)... g=10+3 h>=2 f>=15 child cost=3 h>=2 parent h=5 g=10 h=5 f=15 goal If h is consistent, the f values along any path are nondecreasing. Alistair Knott (Otago) COSC343 Lecture / 1
39 Consistent heuristics Given that A orders the fringe in increasing order of f (n), it follows that A with a consistent heuristic proceeds by successive f-contours, just like breadth-first search proceeds by successive layers. O Z N A I T S R F V L P D M C 420 G B U H E Because of this, we know that the first path we find to a given state will also be the shortest. (And we re safe to throw away re-encountered nodes without checking.) Alistair Knott (Otago) COSC343 Lecture / 1
40 The straight-line distance heuristic is consistent This is easy to show geometrically: parent cost=3 child h>=2 h=5 goal h(child) >= h(parent) cost of path from parent to child Alistair Knott (Otago) COSC343 Lecture / 1
41 Properties of A Complete?? Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
42 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f (G) Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
43 Properties of A Complete?? Yes, unless there are infinitely many nodes with f f (G) Time?? Space?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
44 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?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
45 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?? Optimal?? Alistair Knott (Otago) COSC343 Lecture / 1
46 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?? Alistair Knott (Otago) COSC343 Lecture / 1
47 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?? Alistair Knott (Otago) COSC343 Lecture / 1
48 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 (where C is the path cost of the optimal solution). Alistair Knott (Otago) COSC343 Lecture / 1
49 Some example heuristics for the 8-puzzle Here are a couple of admissible heuristics: The number of misplaced tiles. The total Manhattan distance of each tile from its goal location. (The Manhattan distance between two locations is the number of moves needed to get from one to the other, with no diagonal moves.) Start State Goal State In the above example, misplaced tiles =? Manhattan distance =? Alistair Knott (Otago) COSC343 Lecture / 1
50 Relaxed problems Admissible heuristics can be derived from the exact optimal solution cost of a relaxed version of the problem. If the rules of the 8-puzzle are relaxed so that a tile can move to any location in a single jump, then the number of misplaced tiles is the shortest solution. If the rules are relaxed so that a tile can move to any adjacent square, even if it s already occupied, then total Manhattan distance is 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. So it must be an admissible heuristic. Alistair Knott (Otago) COSC343 Lecture / 1
51 Summary and readings Informed search vs uninformed search Greedy search and A search Admissible and consistent heuristics for guaranteeing optimality of A search For this lecture, you should read AIMA Sections For next lecture: AIMA Sections Alistair Knott (Otago) COSC343 Lecture / 1
Informed search algorithms
Informed search algorithms Chapter 4, Sections 1 2 Chapter 4, Sections 1 2 1 Outline Best-first search A search Heuristics Chapter 4, Sections 1 2 2 Review: Tree search function Tree-Search( problem, fringe)
More informationIntroduction to Artificial Intelligence. Informed Search
Introduction to Artificial Intelligence Informed Search Bernhard Beckert UNIVERSITÄT KOBLENZ-LANDAU Winter Term 2004/2005 B. Beckert: KI für IM p.1 Outline Best-first search A search Heuristics B. Beckert:
More informationInformed search algorithms. Chapter 4, Sections 1 2 1
Informed search algorithms Chapter 4, Sections 1 2 Chapter 4, Sections 1 2 1 Outline Best-first search A search Heuristics Chapter 4, Sections 1 2 2 Review: Tree search function Tree-Search( problem, fringe)
More informationArtificial Intelligence: Search Part 2: Heuristic search
Artificial Intelligence: Search Part 2: Heuristic search Thomas Trappenberg January 16, 2009 Based on the slides provided by Russell and Norvig, Chapter 4, Section 1 2,(4) Outline Best-first search A search
More informationOutline. Informed search algorithms. Best-first search. Review: Tree search. A search Heuristics. Chapter 4, Sections 1 2 4
Outline Best-first search Informed search algorithms A search Heuristics Chapter 4, Sections 1 2 Chapter 4, Sections 1 2 1 Chapter 4, Sections 1 2 2 Review: Tree search function Tree-Search( problem, fringe)
More informationArtificial Intelligence, CS, Nanjing University Spring, 2018, Yang Yu. Lecture 3: Search 2.
Artificial Intelligence, CS, Nanjing University Spring, 2018, Yang Yu Lecture 3: Search 2 http://cs.nju.edu.cn/yuy/course_ai18.ashx Previously... function Tree-Search( problem, fringe) returns a solution,
More informationCS:4420 Artificial Intelligence
CS:4420 Artificial Intelligence Spring 2018 Informed Search Cesare Tinelli The University of Iowa Copyright 2004 18, Cesare Tinelli and Stuart Russell a a These notes were originally developed by Stuart
More informationArtificial Intelligence
Artificial Intelligence hapter 1 hapter 1 1 Iterative deepening search function Iterative-Deepening-Search( problem) returns a solution inputs: problem, a problem for depth 0 to do result Depth-Limited-Search(
More informationInformed search algorithms
CS 580 1 Informed search algorithms Chapter 4, Sections 1 2, 4 CS 580 2 Outline Best-first search A search Heuristics Hill-climbing Simulated annealing CS 580 3 Review: General search function General-Search(
More informationArtificial Intelligence. Informed search methods
Artificial Intelligence Informed search methods In which we see how information about the state space can prevent algorithms from blundering about in the dark. 2 Uninformed vs. Informed Search Uninformed
More informationPlanning, Execution & Learning 1. Heuristic Search Planning
Planning, Execution & Learning 1. Heuristic Search Planning Reid Simmons Planning, Execution & Learning: Heuristic 1 Simmons, Veloso : Fall 2001 Basic Idea Heuristic Search Planning Automatically Analyze
More informationOverview. Path Cost Function. Real Life Problems. COMP219: Artificial Intelligence. Lecture 10: Heuristic Search
COMP219: Artificial Intelligence Lecture 10: Heuristic Search Overview Last time Depth-limited, iterative deepening and bi-directional search Avoiding repeated states Today Show how applying knowledge
More informationCOMP219: Artificial Intelligence. Lecture 10: Heuristic Search
COMP219: Artificial Intelligence Lecture 10: Heuristic Search 1 Class Tests Class Test 1 (Prolog): Tuesday 8 th November (Week 7) 13:00-14:00 LIFS-LT2 and LIFS-LT3 Class Test 2 (Everything but Prolog)
More informationPROBLEM SOLVING AND SEARCH IN ARTIFICIAL INTELLIGENCE
Artificial Intelligence, Computational Logic PROBLEM SOLVING AND SEARCH IN ARTIFICIAL INTELLIGENCE Lecture 3 Informed Search Sarah Gaggl Dresden, 22th April 2014 Agenda 1 Introduction 2 Uninformed Search
More informationTDT4136 Logic and Reasoning Systems
TDT4136 Logic and Reasoning Systems Chapter 3 & 4.1 - Informed Search and Exploration Lester Solbakken solbakke@idi.ntnu.no Norwegian University of Science and Technology 18.10.2011 1 Lester Solbakken
More informationCOMP219: Artificial Intelligence. Lecture 10: Heuristic Search
COMP219: Artificial Intelligence Lecture 10: Heuristic Search 1 Class Tests Class Test 1 (Prolog): Friday 17th November (Week 8), 15:00-17:00. Class Test 2 (Everything but Prolog) Friday 15th December
More informationProblem solving and search
Problem solving and search Chapter 3 Chapter 3 1 Outline Problem-solving agents Problem types Problem formulation Example problems Uninformed search algorithms Informed search algorithms Chapter 3 2 Restricted
More informationProblem solving and search
Problem solving and search Chapter 3 TB Artificial Intelligence Slides from AIMA http://aima.cs.berkeley.edu 1 /1 Outline Problem-solving agents Problem types Problem formulation Example problems Basic
More informationInformed search algorithms
Artificial Intelligence Topic 4 Informed search algorithms Best-first search Greedy search A search Admissible heuristics Memory-bounded search IDA SMA Reading: Russell and Norvig, Chapter 4, Sections
More informationInformed Search and Exploration
Ch. 04 p.1/39 Informed Search and Exploration Chapter 4 Ch. 04 p.2/39 Outline Best-first search A search Heuristics IDA search Hill-climbing Simulated annealing Ch. 04 p.3/39 Review: Tree search function
More informationCOMP9414/ 9814/ 3411: Artificial Intelligence. 5. Informed Search. Russell & Norvig, Chapter 3. UNSW c Alan Blair,
COMP9414/ 9814/ 3411: Artificial Intelligence 5. Informed Search Russell & Norvig, Chapter 3. COMP9414/9814/3411 15s1 Informed Search 1 Search Strategies General Search algorithm: add initial state to
More informationInformed Search and Exploration
Ch. 03 p.1/47 Informed Search and Exploration Sections 3.5 and 3.6 Ch. 03 p.2/47 Outline Best-first search A search Heuristics, pattern databases IDA search (Recursive Best-First Search (RBFS), MA and
More informationLecture Plan. Best-first search Greedy search A* search Designing heuristics. Hill-climbing. 1 Informed search strategies. Informed strategies
Lecture Plan 1 Informed search strategies (KA AGH) 1 czerwca 2010 1 / 28 Blind vs. informed search strategies Blind search methods You already know them: BFS, DFS, UCS et al. They don t analyse the nodes
More informationInformed Search Algorithms. Chapter 4
Informed Search Algorithms Chapter 4 Outline Informed Search and Heuristic Functions For informed search, we use problem-specific knowledge to guide the search. Topics: Best-first search A search Heuristics
More informationInformed Search and Exploration
Ch. 03b p.1/51 Informed Search and Exploration Sections 3.5 and 3.6 Nilufer Onder Department of Computer Science Michigan Technological University Ch. 03b p.2/51 Outline Best-first search A search Heuristics,
More informationRobot Programming with Lisp
6. Search Algorithms Gayane Kazhoyan (Stuart Russell, Peter Norvig) Institute for University of Bremen Contents Problem Definition Uninformed search strategies BFS Uniform-Cost DFS Depth-Limited Iterative
More informationCS414-Artificial Intelligence
CS414-Artificial Intelligence Lecture 6: Informed Search Algorithms Waheed Noor Computer Science and Information Technology, University of Balochistan, Quetta, Pakistan Waheed Noor (CS&IT, UoB, Quetta)
More informationInformed Search. Dr. Richard J. Povinelli. Copyright Richard J. Povinelli Page 1
Informed Search Dr. Richard J. Povinelli Copyright Richard J. Povinelli Page 1 rev 1.1, 9/25/2001 Objectives You should be able to explain and contrast uniformed and informed searches. be able to compare,
More informationIntroduction to Artificial Intelligence (G51IAI)
Introduction to Artificial Intelligence (G51IAI) Dr Rong Qu Heuristic Searches Blind Search vs. Heuristic Searches Blind search Randomly choose where to search in the search tree When problems get large,
More informationInformed Search. Topics. Review: Tree Search. What is Informed Search? Best-First Search
Topics Informed Search Best-First Search Greedy Search A* Search Sattiraju Prabhakar CS771: Classes,, Wichita State University 3/6/2005 AI_InformedSearch 2 Review: Tree Search What is Informed Search?
More informationProblem solving and search
Problem solving and search Chapter 3 Chapter 3 1 How to Solve a (Simple) Problem 7 2 4 1 2 5 6 3 4 5 8 3 1 6 7 8 Start State Goal State Chapter 3 2 Introduction Simple goal-based agents can solve problems
More informationPrincess Nora University Faculty of Computer & Information Systems ARTIFICIAL INTELLIGENCE (CS 370D) Computer Science Department
Princess Nora University Faculty of Computer & Information Systems ARTIFICIAL INTELLIGENCE (CS 370D) Computer Science Department (CHAPTER-3-PART3) PROBLEM SOLVING AND SEARCH Searching algorithm Uninformed
More informationSearching and NetLogo
Searching and NetLogo Artificial Intelligence @ Allegheny College Janyl Jumadinova September 6, 2018 Janyl Jumadinova Searching and NetLogo September 6, 2018 1 / 21 NetLogo NetLogo the Agent Based Modeling
More informationChapter 3: Informed Search and Exploration. Dr. Daisy Tang
Chapter 3: Informed Search and Exploration Dr. Daisy Tang Informed Search Definition: Use problem-specific knowledge beyond the definition of the problem itself Can find solutions more efficiently Best-first
More informationGraphs vs trees up front; use grid too; discuss for BFS, DFS, IDS, UCS Cut back on A* optimality detail; a bit more on importance of heuristics,
Graphs vs trees up front; use grid too; discuss for BFS, DFS, IDS, UCS Cut back on A* optimality detail; a bit more on importance of heuristics, performance data Pattern DBs? General Tree Search function
More informationSolving Problems using Search
Solving Problems using Search Artificial Intelligence @ Allegheny College Janyl Jumadinova September 11, 2018 Janyl Jumadinova Solving Problems using Search September 11, 2018 1 / 35 Example: Romania On
More informationArtificial Intelligence, CS, Nanjing University Spring, 2018, Yang Yu. Lecture 2: Search 1.
rtificial Intelligence, S, Nanjing University Spring, 2018, Yang Yu Lecture 2: Search 1 http://lamda.nju.edu.cn/yuy/course_ai18.ashx Problem in the lecture 7 2 4 51 2 3 5 6 4 5 6 8 3 1 7 8 Start State
More informationCS486/686 Lecture Slides (c) 2015 P.Poupart
1 2 Solving Problems by Searching [RN2] Sec 3.1-3.5 [RN3] Sec 3.1-3.4 CS486/686 University of Waterloo Lecture 2: May 7, 2015 3 Outline Problem solving agents and search Examples Properties of search algorithms
More informationCS486/686 Lecture Slides (c) 2014 P.Poupart
1 2 1 Solving Problems by Searching [RN2] Sec 3.1-3.5 [RN3] Sec 3.1-3.4 CS486/686 University of Waterloo Lecture 2: January 9, 2014 3 Outline Problem solving agents and search Examples Properties of search
More informationSolving Problems by Searching
Solving Problems by Searching CS486/686 University of Waterloo Sept 11, 2008 1 Outline Problem solving agents and search Examples Properties of search algorithms Uninformed search Breadth first Depth first
More informationFoundations of Artificial Intelligence
Foundations of Artificial Intelligence 16. State-Space Search: Greedy BFS, A, Weighted A Malte Helmert University of Basel March 28, 2018 State-Space Search: Overview Chapter overview: state-space search
More informationArtificial Intelligence
Artificial Intelligence Search Marc Toussaint University of Stuttgart Winter 2015/16 (slides based on Stuart Russell s AI course) Outline Problem formulation & examples Basic search algorithms 2/100 Example:
More information16.1 Introduction. Foundations of Artificial Intelligence Introduction Greedy Best-first Search 16.3 A Weighted A. 16.
Foundations of Artificial Intelligence March 28, 2018 16. State-Space Search: Greedy BFS, A, Weighted A Foundations of Artificial Intelligence 16. State-Space Search: Greedy BFS, A, Weighted A Malte Helmert
More information4. Solving Problems by Searching
COMP9414/9814/3411 15s1 Search 1 COMP9414/ 9814/ 3411: Artificial Intelligence 4. Solving Problems by Searching Russell & Norvig, Chapter 3. Motivation Reactive and Model-Based Agents choose their actions
More informationInformed Search and Exploration for Agents
Informed Search and Exploration for Agents R&N: 3.5, 3.6 Michael Rovatsos University of Edinburgh 29 th January 2015 Outline Best-first search Greedy best-first search A * search Heuristics Admissibility
More informationArtificial Intelligence
Artificial Intelligence Academic year 2016/2017 Giorgio Fumera http://pralab.diee.unica.it fumera@diee.unica.it Pattern Recognition and Applications Lab Department of Electrical and Electronic Engineering
More informationArtificial Intelligence CS 6364
Artificial Intelligence CS 6364 Professor Dan Moldovan Section 4 Informed Search and Adversarial Search Outline Best-first search Greedy best-first search A* search Heuristics revisited Minimax search
More informationWeek 3: Path Search. COMP9414/ 9814/ 3411: Artificial Intelligence. Motivation. Example: Romania. Romania Street Map. Russell & Norvig, Chapter 3.
COMP9414/9814/3411 17s1 Search 1 COMP9414/ 9814/ 3411: Artificial Intelligence Week 3: Path Search Russell & Norvig, Chapter 3. Motivation Reactive and Model-Based Agents choose their actions based only
More informationAlgorithm. December 7, Shortest path using A Algorithm. Phaneendhar Reddy Vanam. Introduction. History. Components of A.
December 7, 2011 1 2 3 4 5 6 7 The A is a best-first search algorithm that finds the least cost path from an initial configuration to a final configuration. The most essential part of the A is a good heuristic
More informationFoundations of Artificial Intelligence
Foundations of Artificial Intelligence 4. Informed Search Methods Heuristics, Local Search Methods, Genetic Algorithms Joschka Boedecker and Wolfram Burgard and Frank Hutter and Bernhard Nebel Albert-Ludwigs-Universität
More informationAutomated Planning & Artificial Intelligence
Automated Planning & Artificial Intelligence Uninformed and Informed search in state space Humbert Fiorino Humbert.Fiorino@imag.fr http://membres-lig.imag.fr/fiorino Laboratory of Informatics of Grenoble
More informationProblem solving and search
Problem solving and search hapter 3 hapter 3 1 Outline Problem-solving agents Problem types Problem formulation Example problems asic search algorithms hapter 3 3 Restricted form of general agent: Problem-solving
More informationLars Schmidt-Thieme, Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Germany, Course on Artificial Intelligence,
Course on Artificial Intelligence, winter term 2012/2013 0/25 Artificial Intelligence Artificial Intelligence 2. Informed Search Lars Schmidt-Thieme Information Systems and Machine Learning Lab (ISMLL)
More informationArtificial Intelligence: Search Part 1: Uninformed graph search
rtificial Intelligence: Search Part 1: Uninformed graph search Thomas Trappenberg January 8, 2009 ased on the slides provided by Russell and Norvig, hapter 3 Search outline Part 1: Uninformed search (tree
More informationAGENTS AND ENVIRONMENTS. What is AI in reality?
AGENTS AND ENVIRONMENTS What is AI in reality? AI is our attempt to create a machine that thinks (or acts) humanly (or rationally) Think like a human Cognitive Modeling Think rationally Logic-based Systems
More informationSolving Problems by Searching. Artificial Intelligence Santa Clara University 2016
Solving Problems by Searching Artificial Intelligence Santa Clara University 2016 Problem Solving Agents Problem Solving Agents Use atomic representation of states Planning agents Use factored or structured
More informationInformatics 2D: Tutorial 2 (Solutions)
Informatics 2D: Tutorial 2 (Solutions) Adversarial Search and Informed Search Week 3 1 Adversarial Search This exercise was taken from R&N Chapter 5. Consider the two-player game shown in Figure 1. Figure
More informationAGENTS AND ENVIRONMENTS. What is AI in reality?
AGENTS AND ENVIRONMENTS What is AI in reality? AI is our attempt to create a machine that thinks (or acts) humanly (or rationally) Think like a human Cognitive Modeling Think rationally Logic-based Systems
More informationCSE 473. Chapter 4 Informed Search. CSE AI Faculty. Last Time. Blind Search BFS UC-BFS DFS DLS Iterative Deepening Bidirectional Search
CSE 473 Chapter 4 Informed Search CSE AI Faculty Blind Search BFS UC-BFS DFS DLS Iterative Deepening Bidirectional Search Last Time 2 1 Repeated States Failure to detect repeated states can turn a linear
More informationPEAS: Medical diagnosis system
PEAS: Medical diagnosis system Performance measure Patient health, cost, reputation Environment Patients, medical staff, insurers, courts Actuators Screen display, email Sensors Keyboard/mouse Environment
More informationIntroduction to Artificial Intelligence (G51IAI) Dr Rong Qu. Blind Searches
Introduction to Artificial Intelligence (G51IAI) Dr Rong Qu Blind Searches Blind Searches Function GENERAL-SEARCH (problem, QUEUING-FN) returns a solution or failure nodes = MAKE-QUEUE(MAKE-NODE(INITIAL-STATE[problem]))
More informationOutline for today s lecture. Informed Search I. One issue: How to search backwards? Very briefly: Bidirectional search. Outline for today s lecture
Outline for today s lecture Informed Search I Uninformed Search Briefly: Bidirectional Search (AIMA 3.4.6) Uniform Cost Search (UCS) Informed Search Introduction to Informed search Heuristics 1 st attempt:
More informationOutline. Best-first search
Outline Best-first search Greedy best-first search A* search Heuristics Admissible Heuristics Graph Search Consistent Heuristics Local search algorithms Hill-climbing search Beam search Simulated annealing
More informationCS 380: Artificial Intelligence Lecture #4
CS 380: Artificial Intelligence Lecture #4 William Regli Material Chapter 4 Section 1-3 1 Outline Best-first search Greedy best-first search A * search Heuristics Local search algorithms Hill-climbing
More informationCS:4420 Artificial Intelligence
S:4420 rtificial Intelligence Spring 2018 Uninformed Search esare Tinelli The University of Iowa opyright 2004 18, esare Tinelli and Stuart Russell a a These notes were originally developed by Stuart Russell
More informationArtificial Intelligence. 2. Informed Search
Artificial Intelligence Artificial Intelligence 2. Informed Search Lars Schmidt-Thieme Information Systems and Machine Learning Lab (ISMLL) Institute of Economics and Information Systems & Institute of
More informationInformed Search and Exploration
Informed Search and Exploration Chapter 4 (4.1-4.3) CS 2710 1 Introduction Ch.3 searches good building blocks for learning about search But vastly inefficient eg: Can we do better? Breadth Depth Uniform
More informationOutline. Best-first search
Outline Best-first search Greedy best-first search A* search Heuristics Local search algorithms Hill-climbing search Beam search Simulated annealing search Genetic algorithms Constraint Satisfaction Problems
More informationProblem Solving and Search. Geraint A. Wiggins Professor of Computational Creativity Department of Computer Science Vrije Universiteit Brussel
Problem Solving and Search Geraint A. Wiggins Professor of Computational Creativity Department of Computer Science Vrije Universiteit Brussel What is problem solving? An agent can act by establishing goals
More informationCS 771 Artificial Intelligence. Informed Search
CS 771 Artificial Intelligence Informed Search Outline Review limitations of uninformed search methods Informed (or heuristic) search Uses problem-specific heuristics to improve efficiency Best-first,
More informationCOMP219: Artificial Intelligence. Lecture 7: Search Strategies
COMP219: Artificial Intelligence Lecture 7: Search Strategies 1 Overview Last time basic ideas about problem solving; state space; solutions as paths; the notion of solution cost; the importance of using
More informationSolving Problems by Searching
Solving Problems by Searching Agents, Goal-Based Agents, Problem-Solving Agents Search Problems Blind Search Strategies Agents sensors environment percepts actions? agent effectors Definition. An agent
More informationSolving Problems by Searching
Solving Problems by Searching Agents, Goal-Based Agents, Problem-Solving Agents Search Problems Blind Search Strategies Agents sensors environment percepts actions? agent effectors Definition. An agent
More informationInformed search algorithms. Chapter 4
Informed search algorithms Chapter 4 Material Chapter 4 Section 1 - Exclude memory-bounded heuristic search 3 Outline Best-first search Greedy best-first search A * search Heuristics Local search algorithms
More informationInformed search methods
CS 2710 Foundations of AI Lecture 5 Informed search methods Milos Hauskrecht milos@pitt.edu 5329 Sennott Square Announcements Homework assignment 2 is out Due on Tuesday, September 19, 2017 before the
More informationPlanning and search. Lecture 1: Introduction and Revision of Search. Lecture 1: Introduction and Revision of Search 1
Planning and search Lecture 1: Introduction and Revision of Search Lecture 1: Introduction and Revision of Search 1 Lecturer: Natasha lechina email: nza@cs.nott.ac.uk ontact and web page web page: http://www.cs.nott.ac.uk/
More informationOptimal Control and Dynamic Programming
Optimal Control and Dynamic Programming SC Q 7- Duarte Antunes Outline Shortest paths in graphs Dynamic programming Dijkstra s and A* algorithms Certainty equivalent control Graph Weighted Graph Nodes
More informationInformed search algorithms
Informed search algorithms This lecture topic Chapter 3.5-3.7 Next lecture topic Chapter 4.1-4.2 (Please read lecture topic material before and after each lecture on that topic) Outline Review limitations
More informationOutline. Robotics Planning. Example: Getting to DC from Atlanta. Introduction
Outline Robotics Planning Henrik I Christensen Introduction Representations Planning Basics Search methods Evaluation (or the lack thereof) Summary Introduction Example: Getting to DC from Atlanta Generation
More informationInformed search algorithms
Informed search algorithms This lecture topic Chapter 3.5-3.7 Next lecture topic Chapter 4.1-4.2 (Please read lecture topic material before and after each lecture on that topic) Outline Review limitations
More informationSummary. Search CSL302 - ARTIFICIAL INTELLIGENCE 1
Summary Search CSL302 - ARTIFICIAL INTELLIGENCE 1 Informed Search CHAPTER 3 Informed (Heuristic) Search qheuristic problem-specific knowledge ofinds solutions more efficiently qnew terms o!(#): cost from
More informationClustering (Un-supervised Learning)
Clustering (Un-supervised Learning) Partition-based clustering: k-mean Goal: minimize sum of square of distance o Between each point and centers of the cluster. o Between each pair of points in the cluster
More informationInformed Search. Best-first search. Greedy best-first search. Intelligent Systems and HCI D7023E. Romania with step costs in km
Informed Search Intelligent Systems and HCI D7023E Lecture 5: Informed Search (heuristics) Paweł Pietrzak [Sec 3.5-3.6,Ch.4] A search strategy which searches the most promising branches of the state-space
More informationA.I.: Informed Search Algorithms. Chapter III: Part Deux
A.I.: Informed Search Algorithms Chapter III: Part Deux Best-first search Greedy best-first search A * search Heuristics Outline Overview Informed Search: uses problem-specific knowledge. General approach:
More informationArtificial Intelligence
Artificial Intelligence CSC348 Unit 3: Problem Solving and Search Syedur Rahman Lecturer, CSE Department North South University syedur.rahman@wolfson.oxon.org Artificial Intelligence: Lecture Notes The
More informationSolving problems by searching
Solving problems by searching Chapter 3 Some slide credits to Hwee Tou Ng (Singapore) Outline Problem-solving agents Problem types Problem formulation Example problems Basic search algorithms Heuristics
More informationSeminar: Search and Optimization
Seminar: Search and Optimization 4. asic Search lgorithms Martin Wehrle Universität asel October 4, 2012 asics lind Search lgorithms est-first Search Summary asics asics lind Search lgorithms est-first
More informationPart I. Instructor: Dr. Wei Ding. Uninformed Search Strategies can find solutions to problems by. Informed Search Strategies
Informed Search and Exploration Part I Instructor: Dr. Wei Ding Fall 2010 1 Motivation Uninformed Search Strategies can find solutions to problems by Systematically generating new states Testing them against
More informationARTIFICIAL INTELLIGENCE. Informed search
INFOB2KI 2017-2018 Utrecht University The Netherlands ARTIFICIAL INTELLIGENCE Informed search Lecturer: Silja Renooij These slides are part of the INFOB2KI Course Notes available from www.cs.uu.nl/docs/vakken/b2ki/schema.html
More informationTDDC17. Intuitions behind heuristic search. Recall Uniform-Cost Search. Best-First Search. f(n) =... + h(n) g(n) = cost of path from root node to n
Intuitions behind heuristic search The separation property of GRAPH-SEARCH TDDC17 Seminar III Search II Informed or Heuristic Search Beyond Classical Search Find a heuristic measure h(n) which estimates
More informationCS-E4800 Artificial Intelligence
CS-E4800 Artificial Intelligence Jussi Rintanen Department of Computer Science Aalto University January 12, 2017 Transition System Models The decision-making and planning at the top-level of many intelligent
More informationShortest path using A Algorithm
Shortest path using A Algorithm Phaneendhar Reddy Vanam Computer Science Indiana State University Terre Haute, IN, USA December 13, 2011 Contents Abstract The main aim of this project is to find the shortest
More informationInformed Search. Notes about the assignment. Outline. Tree search: Reminder. Heuristics. Best-first search. Russell and Norvig chap.
Notes about the assignment Informed Search Russell and Norvig chap. 4 If it says return True or False, return True or False, not "True" or "False Comment out or remove print statements before submitting.
More informationLecture 4: Informed/Heuristic Search
Lecture 4: Informed/Heuristic Search Outline Limitations of uninformed search methods Informed (or heuristic) search uses problem-specific heuristics to improve efficiency Best-first A* RBFS SMA* Techniques
More informationChapter 3 Solving Problems by Searching Informed (heuristic) search strategies More on heuristics
Chapter 3 Solving Problems by Searching 3.5 3.6 Informed (heuristic) search strategies More on heuristics CS5811 - Advanced Artificial Intelligence Nilufer Onder Department of Computer Science Michigan
More informationEE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS
EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS Lecture 4, 4/11/2005 University of Washington, Department of Electrical Engineering Spring 2005 Instructor: Professor Jeff A. Bilmes Today: Informed search algorithms
More informationFoundations of Artificial Intelligence
Foundations of Artificial Intelligence 4. Informed Search Methods Heuristics, Local Search Methods, Genetic Algorithms Joschka Boedecker and Wolfram Burgard and Bernhard Nebel Albert-Ludwigs-Universität
More informationProblem-solving agents. Solving Problems by Searching. Outline. Example: Romania. Chapter 3
Problem-solving agents olving Problems by earching hapter 3 function imple-problem-olving-gent( percept) returns an action static: seq, an action sequence, initially empty state, some description of the
More informationCS 5100: Founda.ons of Ar.ficial Intelligence
CS 5100: Founda.ons of Ar.ficial Intelligence Search Problems and Solutions Prof. Amy Sliva October 6, 2011 Outline Review inference in FOL Most general unidiers Conversion to CNF UniDication algorithm
More informationSearch. Intelligent agents. Problem-solving. Problem-solving agents. Road map of Romania. The components of a problem. that will take me to the goal!
Search Intelligent agents Reflex agent Problem-solving agent T65 rtificial intelligence and Lisp Peter alenius petda@ida.liu.se epartment of omputer and Information Science Linköping University Percept
More information