Download presentation
Presentation is loading. Please wait.
Published byErika Holmes Modified over 9 years ago
1
1 CSE 4705 Artificial Intelligence Jinbo Bi Department of Computer Science & Engineering http://www.engr.uconn.edu/~jinbo
2
2 TodayToday Intelligent Agents
3
3 Inverted pendulum Example to demonstrate a learning agent
4
4 8-puzzle8-puzzle A tile adjacent to the blank space can slide into the space.
5
5 Holiday in Romania Start Goal
6
6 Complexity of Breadth-First Search
7
7 Holiday in Romania Start Goal
8
8 ComparisonComparison
9
9 Demonstration on Games/Robots Breadth First Search Pink: starting point Blue: goal Teal: scanned squares Darker: closer to starting point
10
10 Demonstration on Games/Robots An optimal informed search algorithm A* We add a heuristic estimate of distance to the goal Yellow: examined nodes with high h(n) Blue: examined nodes with low h(n)
11
11 Demonstration on Games/Robots Breadth-first search expands many many nodes Pink: starting node Dark blue: goal
12
12 Demonstration on Games/Robots A* search expands much fewer nodes Pink: starting node Dark blue: goal
13
13 Start Goal The distance from each city to Bucharest:
14
14 Best-first Search
15
15 Best-first Search
16
16 A* Search
17
17 A* Search
18
18 A* Search
19
19 Hill Climbing
20
20 8-puzzle8-puzzle Start Goal
21
21 Hill-Climbing Ex: 8-queens
22
22 Gradient ascent/descent
23
23 Gradient methods / Newton’s methods Contour lines of a function (Green: gradient descent, Red: Newton’s methods)
24
24 Difficult Problems
25
25 Difficult Problems
26
26 Random Restart
27
27 Genetic Algorithm https://www.youtube.com/watch?v=ejxfTy4lI6I A short video explains Genetic Algorithm in 3 minutes
28
28 Genetic Algorithm
29
29 Searching nondeterministic The 8 physical states of the vacuum world
30
30 Searching nondeterministic Fig. 4.10, AND-OR Search Tree, and a depth-first search
31
31 Searching nondeterministic Fig. 4.11, AND-OR Search algorithm (graph search) and a depth-first search, it returns a conditional plan that reaches a goal state in all circumstances S i in
32
32 Searching partial observable Deterministic Non-deterministic Fig. 4.13
33
33 Searching partial observable
34
34 Searching partial observable A vacuum has local sensors, and can report a state of [location, dirty/clean]
35
35 Searching partial observable Partial observations can still be quite useful (Fig. 4.18
36
36 Game Tree for Tic-Tac-Toe
37
37 An Evaluation Function for Tic-Tac-Toe f(n) = 8-8=0 f(n) = 8-5=3 f(n) = 8-6=2 f(n) = 2f(n) = 3 f(n): the potential # of lines with 3 x – the potential # of lines with three o f(n) = 0 if a tie f(n) = + ∞ if n is a terminal win f(n) = - ∞ if n is a terminal loss
38
38 Two Players MINIMAX value for a Two-Play Game Tree
39
39 Multiple Players
40
40 Alpha-Beta Pruning
41
41 Map Coloring
42
42 A Consistent and Complete Solution to Map Coloring
43
43 BacktrackingBacktracking
44
44 Backtracking – Map Coloring
45
45 Improving Backtracking Most constrained variables Most constraining variables
46
46 Improving Backtracking Given n variables, choose the least constraining value
47
47 Improving Backtracking Forward checking
48
48 Arc Consistency: General Case If X-> Y is consistent iff for every value x of X there is some allowed y in Y that can be used
49
49 Arc Consistency
50
50 ≠ General Backtracking
51
51 The Wumpus World http://www.flashrolls.com/puzzle-games/Hunt-The-Wumpus- Flash-Game.htm
52
52 The Wumpus World Figure 7.3 The first step by the agent in the wumpus world
53
53 The Wumpus World Figure 7.4 The second and third steps by the agent in the wumpus world
54
54 The Wumpus World
55
55 The Wumpus World Figure 7.5 Dotted line shows the model of (no pit in [1,2])
56
56 The Wumpus World Figure 7.5 Dotted line shows the model of (no pit in [2,2])
57
57 Truth Table
58
58 Logical Equivalence
59
59 Resolution Algorithm To prove KB entails α, we need to show KB -> α is valid. By proof of contradiction, we need to prove KB ˄ ~ α is unsatisfiable, which means the resolution algorithm will give clauses including an empty clause.
60
60 Resolution Algorithm The Wumpus example
61
61 Forward Chaining
62
62 Forward Chaining (ex)
63
63 Forward Chaining (ex)
64
64 The WalkSAT Algorithm
65
65 Hard Satisfiability Problems
66
66 Hard Satisfiability Problems Median runtime for 100 satisfiable random 3-CNF statements, n=50
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.