searchcs188/su20/assets/lecture/lec2-su20.pdf§lecture 1 slides and recording § see piazza. cs 188:...
TRANSCRIPT
![Page 1: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/1.jpg)
CS 188: Artificial Intelligence
Search
Instructor: Nikita Kitaev
University of California, Berkeley[These slides adapted from Dan Klein and Pieter Abbeel]
![Page 2: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/2.jpg)
Announcements§ Exam Times
§ Midterm 1: July 13, 12pm-2pm (Alternative 12 hours later at 12am-2am)§ Midterm 2: July 29, 12pm-2pm (Alternative 12 hours later at 12am-2am)§ Final: August 13, 12pm-3pm (Alternative 12 hours later at 12am-3am)
§ Sections§ Start tomorrow, see Piazza for times§ You can go to any section
§ Office Hours§ Start today§ Ryan’s OH are today at 3pm-5pm. Cathy’s OH today at 11pm-12am. Nathan’s OH tomorrow at 7am-9am
§ Written Assessment 1 will be released tonight§ due next Monday before start of lecture
§ Lecture 1 slides and recording§ See Piazza
![Page 3: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/3.jpg)
CS 188: Artificial Intelligence
Search
Instructor: Nikita Kitaev
University of California, Berkeley[These slides adapted from Dan Klein and Pieter Abbeel]
![Page 4: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/4.jpg)
Today
§ Agents that Plan Ahead
§ Search Problems
§ Uninformed Search Methods§ Depth-First Search§ Breadth-First Search§ Uniform-Cost Search
![Page 5: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/5.jpg)
Agents that Plan
![Page 6: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/6.jpg)
Reflex Agents
§ Reflex agents:§ Choose action based on current percept (and
maybe memory)§ May have memory or a model of the world’s
current state§ Do not consider the future consequences of
their actions§ Consider how the world IS
§ Can a reflex agent be rational?
![Page 7: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/7.jpg)
Video of Demo Reflex Optimal
![Page 8: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/8.jpg)
Video of Demo Reflex Odd
![Page 9: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/9.jpg)
Planning Agents
§ Planning agents:§ Ask “what if”§ Decisions based on (hypothesized)
consequences of actions§ Must have a model of how the world evolves in
response to actions§ Must formulate a goal (test)§ Consider how the world WOULD BE
§ Optimal vs. complete planning
§ Planning vs. replanning
![Page 10: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/10.jpg)
Video of Demo Replanning
![Page 11: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/11.jpg)
Video of Demo Mastermind
![Page 12: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/12.jpg)
Search Problems
![Page 13: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/13.jpg)
Search Problems
§ A search problem consists of:
§ A state space
§ A successor function(with actions, costs)
§ A start state and a goal test
§ A solution is a sequence of actions (a plan) which transforms the start state to a goal state
“N”, 1.0
“E”, 1.0
![Page 14: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/14.jpg)
Search Problems Are Models
![Page 15: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/15.jpg)
Example: Traveling in Romania
§ State space:§ Cities
§ Successor function:§ Roads: Go to adjacent city with
cost = distance
§ Start state:§ Arad
§ Goal test:§ Is state == Bucharest?
§ Solution?
![Page 16: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/16.jpg)
What’s in a State Space?
§ Problem: Pathing§ States: (x,y) location§ Actions: NSEW§ Successor: update location
only§ Goal test: is (x,y)=END
§ Problem: Eat-All-Dots§ States: {(x,y), dot booleans}§ Actions: NSEW§ Successor: update location
and possibly a dot boolean§ Goal test: dots all false
The world state includes every last detail of the environment
A search state keeps only the details needed for planning (abstraction)
![Page 17: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/17.jpg)
State Space Sizes?
§ World state:§ Agent positions: 120§ Food count: 30§ Ghost positions: 12§ Agent facing: NSEW
§ How many§ World states?
120x(230)x(122)x4§ States for pathing?
120§ States for eat-all-dots?
120x(230)
![Page 18: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/18.jpg)
Quiz: Safe Passage
§ Problem: eat all dots while keeping the ghosts perma-scared§ What does the state space have to specify?
§ (agent position, dot booleans, power pellet booleans, remaining scared time)
![Page 19: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/19.jpg)
State Space Graphs and Search Trees
![Page 20: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/20.jpg)
State Space Graphs
§ State space graph: A mathematical representation of a search problem§ Nodes are (abstracted) world configurations§ Arcs represent successors (action results)§ The goal test is a set of goal nodes (maybe only one)
§ In a state space graph, each state occurs only once!
§ We can rarely build this full graph in memory (it’s too big), but it’s a useful idea
![Page 21: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/21.jpg)
State Space Graphs
§ State space graph: A mathematical representation of a search problem§ Nodes are (abstracted) world configurations§ Arcs represent successors (action results)§ The goal test is a set of goal nodes (maybe only one)
§ In a state space graph, each state occurs only once!
§ We can rarely build this full graph in memory (it’s too big), but it’s a useful idea
S
G
d
b
p q
c
e
h
a
f
r
Tiny state space graph for a tiny search problem
![Page 22: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/22.jpg)
Search Trees
§ A search tree:§ A “what if” tree of plans and their outcomes§ The start state is the root node§ Children correspond to successors§ Nodes show states, but correspond to PLANS that achieve those states§ For most problems, we can never actually build the whole tree
“E”, 1.0“N”, 1.0
This is now / start
Possible futures
![Page 23: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/23.jpg)
State Space Graphs vs. Search Trees
S
a
b
d p
a
c
e
p
h
f
r
q
q c G
a
qe
p
h
f
r
q
q c Ga
S
G
d
b
p q
c
e
h
a
f
r
We construct both on demand – and we construct as little as possible.
Each NODE in in the search tree is an entire PATH in the state space
graph.
Search TreeState Space Graph
![Page 24: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/24.jpg)
Quiz: State Space Graphs vs. Search Trees
S G
b
a
Consider this 4-state graph: How big is its search tree (from S)?
![Page 25: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/25.jpg)
Quiz: State Space Graphs vs. Search Trees
S G
b
a
Consider this 4-state graph:
Important: Lots of repeated structure in the search tree!
How big is its search tree (from S)?
sb
b G a
a
G
a G b G
… …
![Page 26: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/26.jpg)
Tree Search
![Page 27: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/27.jpg)
Search Example: Romania
![Page 28: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/28.jpg)
Searching with a Search Tree
§ Search:§ Expand out potential plans (tree nodes)§ Maintain a fringe of partial plans under consideration§ Try to expand as few tree nodes as possible
![Page 29: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/29.jpg)
General Tree Search
§ Important ideas:§ Fringe§ Expansion§ Exploration strategy
§ Main question: which fringe nodes to explore?
![Page 30: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/30.jpg)
Example: Tree Search
S
G
d
b
p q
c
e
h
a
f
r
![Page 31: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/31.jpg)
Example: Tree Search
a a p
q
h
f
r
q
c G
a
q
qp
q
a
S
G
d
b
p q
c
e
h
a
f
r
fd e
r
S
d e p
e
h r
f
c G
b c
ss à ds à es à ps à d à bs à d à cs à d à es à d à e à hs à d à e à rs à d à e à r à fs à d à e à r à f à cs à d à e à r à f à G
![Page 32: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/32.jpg)
Depth-First Search
![Page 33: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/33.jpg)
Depth-First Search
S
a
b
d p
a
c
e
p
h
f
r
q
q c G
a
qe
p
h
f
r
q
q c G
a
S
G
d
b
p q
c
e
h
a
f
rqph
fd
ba
c
e
r
Strategy: expand a deepest node first
Implementation: Fringe is a LIFO stack
![Page 34: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/34.jpg)
Search Algorithm Properties
![Page 35: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/35.jpg)
Search Algorithm Properties
§ Complete: Guaranteed to find a solution if one exists?§ Optimal: Guaranteed to find the least cost path?§ Time complexity?§ Space complexity?
§ Cartoon of search tree:§ b is the branching factor§ m is the maximum depth§ solutions at various depths
§ Number of nodes in entire tree?§ 1 + b + b2 + …. bm = O(bm)
…b 1 node
b nodes
b2 nodes
bm nodes
m tiers
![Page 36: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/36.jpg)
Depth-First Search (DFS) Properties
…b 1 node
b nodes
b2 nodes
bm nodes
m tiers
§ What nodes DFS expand?§ Some left prefix of the tree.§ Could process the whole tree!§ If m is finite, takes time O(bm)
§ How much space does the fringe take?§ Only has siblings on path to root, so O(bm)
§ Is it complete?§ m could be infinite, so only if we prevent
cycles (more later)
§ Is it optimal?§ No, it finds the “leftmost” solution,
regardless of depth or cost
![Page 37: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/37.jpg)
Breadth-First Search
![Page 38: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/38.jpg)
Breadth-First Search
S
a
b
d p
a
c
e
p
h
f
r
q
q c G
a
qe
p
h
f
r
q
q c G
a
S
G
d
b
p q
ce
h
a
f
r
Search
Tiers
Strategy: expand a shallowest node first
Implementation: Fringe is a FIFO queue
![Page 39: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/39.jpg)
Breadth-First Search (BFS) Properties
§ What nodes does BFS expand?§ Processes all nodes above shallowest solution§ Let depth of shallowest solution be s§ Search takes time O(bs)
§ How much space does the fringe take?§ Has roughly the last tier, so O(bs)
§ Is it complete?§ s must be finite if a solution exists, so yes!
§ Is it optimal?§ Only if costs are all 1 (more on costs later)
…b 1 node
b nodes
b2 nodes
bm nodes
s tiers
bs nodes
![Page 40: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/40.jpg)
Quiz: DFS vs BFS
![Page 41: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/41.jpg)
Video of Demo Maze Water DFS/BFS (part 1)
![Page 42: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/42.jpg)
Video of Demo Maze Water DFS/BFS (part 2)
![Page 43: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/43.jpg)
Quiz: DFS vs BFS
§ When will BFS outperform DFS?
§ When will DFS outperform BFS?
![Page 44: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/44.jpg)
Iterative Deepening
…b
§ Idea: get DFS’s space advantage with BFS’s time / shallow-solution advantages§ Run a DFS with depth limit 1. If no solution…§ Run a DFS with depth limit 2. If no solution…§ Run a DFS with depth limit 3. …..
§ Isn’t that wastefully redundant?§ Generally most work happens in the lowest
level searched, so not so bad!
![Page 45: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/45.jpg)
Cost-Sensitive Search
BFS finds the shortest path in terms of number of actions.It does not find the least-cost path. We will now covera similar algorithm which does find the least-cost path.
START
GOAL
d
b
pq
c
e
h
a
f
r
2
9 2
81
8
2
3
2
4
4
15
1
32
2
![Page 46: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/46.jpg)
Uniform Cost Search
![Page 47: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/47.jpg)
Uniform Cost Search
S
a
b
d p
a
c
e
p
h
f
r
q
q c G
a
qe
p
h
f
r
q
q c G
a
Strategy: expand a cheapest node first:
Fringe is a priority queue (priority: cumulative cost) S
G
d
b
p q
c
e
h
a
f
r
3 9 1
164 115
713
8
1011
17 11
0
6
3 9
1
1
2
8
8 2
15
1
2
Cost contours
2
![Page 48: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/48.jpg)
…
Uniform Cost Search (UCS) Properties
§ What nodes does UCS expand?§ Processes all nodes with cost less than cheapest solution!§ If that solution costs C* and arcs cost at least e , then the
“effective depth” is roughly C*/e§ Takes time O(bC*/e) (exponential in effective depth)
§ How much space does the fringe take?§ Has roughly the last tier, so O(bC*/e)
§ Is it complete?§ Assuming best solution has a finite cost and minimum arc cost
is positive, yes!
§ Is it optimal?§ Yes! (Proof next lecture via A*)
b
C*/e “tiers”c £ 3
c £ 2c £ 1
![Page 49: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/49.jpg)
Uniform Cost Issues
§ Remember: UCS explores increasing cost contours
§ The good: UCS is complete and optimal!
§ The bad:§ Explores options in every “direction”§ No information about goal location
§ We’ll fix that soon!
Start Goal
…
c £ 3c £ 2
c £ 1
![Page 50: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/50.jpg)
Video of Demo Empty UCS
![Page 51: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/51.jpg)
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part 1)
![Page 52: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/52.jpg)
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part 2)
![Page 53: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/53.jpg)
Video of Demo Maze with Deep/Shallow Water --- DFS, BFS, or UCS? (part 3)
![Page 54: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/54.jpg)
The One Queue
§ All these search algorithms are the same except for fringe strategies§ Conceptually, all fringes are priority
queues (i.e. collections of nodes with attached priorities)
§ Practically, for DFS and BFS, you can avoid the log(n) overhead from an actual priority queue, by using stacks and queues
§ Can even code one implementation that takes a variable queuing object
![Page 55: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/55.jpg)
Search and Models
§ Search operates over models of the world§ The agent doesn’t
actually try all the plans out in the real world!
§ Planning is all “in simulation”
§ Your search is only as good as your models…
![Page 56: Searchcs188/su20/assets/lecture/lec2-su20.pdf§Lecture 1 slides and recording § See Piazza. CS 188: Artificial Intelligence Search Instructor: Nikita Kitaev University of California,](https://reader036.vdocuments.net/reader036/viewer/2022081406/5f11aed88e020e0ff520a827/html5/thumbnails/56.jpg)
Search Gone Wrong?