DS/CMPSC 442: Artificial Intelligence

Search

Instructor: Amulya Yadav

Penn State University

1
 Today
o Agents that Plan Ahead

o Search Problems

o Uninformed Search Methods

o Depth-First Search
o Breadth-First Search
o Uniform-Cost Search

2
Why Search Problems?

3
Why Search Problems?

THIS IS THE TOOL OF SEARCH PROBLEMS

4
 Why Search Problems?
o Let’s say there is a real world problem that needs to be
solved.
o Goal: Build an AI agent to solve this problem.
o How to build this agent?

5
 Why Search Problems?
o How to build this AI agent?

o Maybe the problem is simple enough that the answer is

straightforward.

6
 Why Search Problems?
o More interesting problems require the framework of search
problems
o more than one possible alternative needs to be explored before the
problem is solved
o the number of alternatives to search among can be very large

o Can we model the real-world problem using a search problem?

o Develop an AI agent to solve that search problem.

7
Some Characteristics of When to use Search Problems

o Single Agent

o Completely Observable Environment

o No adversaries (single player)

8
 Reflex Agents

o Reflex agents:
o Choose action based on current percept
o May have memory
o Do not consider the future consequences of
their actions
o Consider how the world IS

o Can a reflex agent be rational?

[Demo: reflex optimal (L2D1)]

[Demo: reflex optimal (L2D2)]
9
Video of Demo Reflex Optimal

10
Video of Demo Reflex Odd

11
 Planning Agents

o Planning agents:
o Ask “what if”
o Decisions based on (hypothesized)
consequences of actions
o Must have a model of how the world evolves
in response to actions
o Must formulate a goal (test)
o Consider how the world WOULD BE

o Optimal vs. complete planning

o Planning vs. replanning

[Demo: re-planning (L2D3)]
[Demo: mastermind (L2D4)]
12
Video of Demo Replanning

13
Video of Demo Mastermind

14
Search Problems

15
 Search Problems
o A search problem consists of:

o A state space

“N”, 1.0
o A successor function
(with actions, costs)

“E”, 1.0
o A start state and a goal test

o A solution is a sequence of actions (a plan) which

transforms the start state to a goal state

16
Search Problems Are Models

17
Search – Not Just Pathing Problems

18
Search – Not Just Pathing Problems

19
Search – Not Just Pathing Problems

20
Search – Not Just Pathing Problems

21
Example: Traveling in Romania

o State space:
o Cities
o Successor function:
o Roads: Go to adjacent city with
cost = distance
o Start state:
o Arad
o Goal test:
o Is state == Bucharest?

o Solution?

22
 What’s in a State Space?
The world state includes every last detail of the environment

A search state keeps only the details needed for planning (abstraction)

o Problem: Pathing o Problem: Eat-All-Dots

o States: (x,y) location o States: {(x,y), dot booleans}
o Actions: NSEW o Actions: NSEW
o Successor: update location o Successor: update location
only and possibly a dot boolean
o Goal test: is (x,y)=END o Goal test: dots all false

23
 State Space Sizes?
o World state:
o Agent positions: 120
o Food count: 30
o Ghost positions: 12
o Agent facing: NSEW

o How many
o World states?
120x(230)x(122)x4
o States for pathing?
120
o States for eat-all-dots?
120x(230)

24
 Safe Passage

o Problem: eat all dots while keeping the ghosts perma-scared

o What does the state space have to specify?
o (agent position, dot booleans, power pellet booleans, remaining scared time)

25
State Space Graphs and Search Trees

26
 State Space Graphs

o State space graph: A mathematical

representation of a search problem
o Nodes are (abstracted) world configurations
o Arcs represent successors (action results)
o The goal test is a set of goal nodes (maybe only one)

o In a state space graph, each state occurs only

once!

o We can rarely build this full graph in

memory (it’s too big), but it’s a useful idea

27
 State Space Graphs

o State space graph: A mathematical

a G
representation of a search problem
b c
o Nodes are (abstracted) world configurations
o Arcs represent successors (action results) e
d f
o The goal test is a set of goal nodes (maybe only one)
S h
p r
o In a state space graph, each state occurs only q
once!
Tiny search graph for a tiny
search problem
o We can rarely build this full graph in
memory (it’s too big), but it’s a useful idea

28
 Search Trees
This is now / start
“N”, 1.0 “E”, 1.0

Possible futures

o A search tree:
o A “what if” tree of plans and their outcomes
o The start state is the root node
o Children correspond to successors
o Nodes show states, but correspond to PLANS that achieve those states
o For most problems, we can never actually build the whole tree

29
 State Space Graphs vs. Search Trees

Each NODE in in
State Space Graph the search tree is Search Tree
an entire PATH in
S
the state space
a G graph. d e p
b c
b c e h r q
e
d f h r p q f
a a
S h We construct both
on demand – and p q f q c G
p q r
we construct as q c G a
little as possible.
a

30
 State Space Graphs vs. Search Trees

Consider this 4-state graph: How big is its search tree (from S)?

S G

31
 State Space Graphs vs. Search Trees

Consider this 4-state graph: How big is its search tree (from S)?

a s
a b
S G
b G a G
b a G b G

… …

Important: Lots of repeated structure in the search tree!

32
Tree Search

33
Search Example: Romania

34
 Searching with a Search Tree

o Search:
o Expand out potential plans (tree nodes)
o Maintain a fringe of partial plans under consideration
o Try to expand as few tree nodes as possible

35
 General Tree Search

o Important ideas:
o Fringe
o Expansion
o Exploration strategy

o Main question: which fringe nodes to explore?

36
Example: Tree Search
a G
b c
e
d f
S h
p q r

37
 Example: Tree Search
a G
b c
e
d f
S h
p q r

S s
s d
d e p s e
q s p
b c e h r
s d b
a a h r p q f s d c
s d e
p q f q c G s d e h
a s d e r
q c G
s d e r f
a s d e r f c
38
s d e r f G
Depth-First Search

39
 Depth-First Search
Strategy: expand a a G
deepest node first b c

Implementation: e
d f
Fringe is a LIFO stack S h
p q r

d e p

b c e h r q

a a h r p q f

p q f q c G

q c G a

a
40
Search Algorithm Properties

41
 Search Algorithm Properties
o Complete: Guaranteed to find a solution if one exists?
o Optimal: Guaranteed to find the least cost path?
o Time complexity?
o Space complexity? 1 node
b
… b nodes
o Cartoon of search tree: b2 nodes
o b is the branching factor m tiers
o m is the maximum depth
o solutions at various depths

bm nodes
o Number of nodes in entire tree?
o 1 + b + b2 + …. bm = O(bm)

42
 Depth-First Search (DFS) Properties
o What nodes DFS expand?
o Some left prefix of the tree. 1 node
o Could process the whole tree! b
… b nodes
o If m is finite, takes time O(bm)
b2 nodes
o How much space does the fringe take? m tiers
o Only has siblings on path to root, so O(bm)

o Is it complete?
o m could be infinite, so only if we prevent cycles (more bm nodes
later)

o Is it optimal?
o No, it finds the “leftmost” solution, regardless of depth
or cost

43
Breadth-First Search

44
 Breadth-First Search
Strategy: expand a a G
shallowest node first b c
e
Implementation: Fringe d f
is a FIFO queue S h
p q r

d e p
Search
b c e h r q
Tiers
a a h r p q f

p q f q c G

q c G a

a
45
 Breadth-First Search (BFS) Properties
o What nodes does BFS expand?
o Processes all nodes above shallowest solution 1 node
b
o Let depth of shallowest solution be s … b nodes
s tiers
o Search takes time O(bs) b2 nodes

o How much space does the fringe take? bs nodes

o Has roughly the last tier, so O(bs)

o Is it complete? bm nodes

o s must be finite if a solution exists, so yes!

o Is it optimal?
o Only if costs are all 1 (more on costs later)

46
Quiz: DFS vs BFS

47
 DFS vs BFS

o When will BFS outperform DFS?

o When will DFS outperform BFS?

[Demo: dfs/bfs maze water (L2D6)]

48
Video of Demo Maze Water DFS/BFS (part 1)

49
Video of Demo Maze Water DFS/BFS (part 2)

50
 Iterative Deepening
o Idea: get DFS’s space advantage with
BFS’s time / shallow-solution advantages b
o Run a DFS with depth limit 1. If no solution… …

o Run a DFS with depth limit 2. If no solution…

o Run a DFS with depth limit 3. …..

o Isn’t that wastefully redundant?

o Generally most work happens in the lowest
level searched, so not so bad!

51
 Cost-Sensitive Search
a GOAL
2 2
b c
3
2
1 8
2 e
3 d
f
9 8 2
START h
1 4 2

p 4 r
15
q

BFS finds the shortest path in terms of number of actions.

It does not find the least-cost path. We will now cover How?
a similar algorithm which does find the least-cost path.
52
Uniform Cost Search

53
 Uniform Cost Search
2 a G
Strategy: expand a b c
cheapest node first: 1 8 2
2 e
3 d f
Fringe is a priority queue 9 2
S h 8 1
(priority: cumulative cost)
1 p q r
15

S 0

d 3 e 9 p 1

b 4 c e 5 h 17 r 11 q 16
11
Cost a 6 a h 13 r 7 p q f
contours
p q f 8 q c G

q 11 c G 10 a

54
 Uniform Cost Search (UCS) Properties
o What nodes does UCS expand?
o Processes all nodes with cost less than cheapest solution!
o If that solution costs C* and arcs cost at least ε , then the “effective b c≤1
…
depth” is roughly C*/ε c≤2
o Takes time O(bC*/ε) (exponential in effective depth) C*/ε “tiers”
c≤3
o How much space does the fringe take?
o Has roughly the last tier, so O(bC*/ε)

o Is it complete?
o Assuming best solution has a finite cost and minimum arc cost is
positive, yes!

o Is it optimal?
o Yes! (Proof next lecture via A*)

55
 Uniform Cost Issues
o Remember: UCS explores increasing cost c≤1
…
contours c≤2
c≤3

o The good: UCS is complete and optimal!

o The bad:
o Explores options in every “direction”
o No information about goal location

Start Goal

o We’ll fix that soon!

[Demo: empty grid UCS (L2D5)]
[Demo: maze with deep/shallow
water DFS/BFS/UCS (L2D7)]
56
Video of Demo Empty UCS

57
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part
1)

58
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part
2)

59
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part
3)

60
 The One Queue
o All these search algorithms are the
same except for fringe strategies
o Conceptually, all fringes are priority
queues (i.e. collections of nodes with
attached priorities)
o Practically, for DFS and BFS, you can
avoid the log(n) overhead from an
actual priority queue, by using stacks
and queues
o Can even code one implementation
that takes a variable queuing object

61
 Search and Models

o Search operates over

models of the world
o The agent doesn’t
actually try all the plans
out in the real world!
o Planning is all “in
simulation”
o Your search is only as
good as your models…

62
Search Gone Wrong?

63