Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern...
Transcript of Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern...
![Page 1: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/1.jpg)
Heuristic Search
Robert PlattNortheastern University
Some images and slides are used from:1. CS188 UC Berkeley2. RN, AIMA
![Page 2: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/2.jpg)
Recap: What is graph search?
Graph search: find a path from start to goal
– what are the states?
– what are the actions (transitions)?
– how is this a graph?
Start state Goal state
![Page 3: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/3.jpg)
Recap: What is graph search?
Graph search: find a path from start to goal
– what are the states?
– what are the actions (transitions)?
– how is this a graph?
Start state
Goal state
![Page 4: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/4.jpg)
Recap: BFS/UCS
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 5: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/5.jpg)
Recap: BFS/UCS
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
Start GoalIt's like this
Notice that we search equally far in all directions...
![Page 6: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/6.jpg)
Idea
Is it possible to use additional information to decide which direction to search in?
![Page 7: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/7.jpg)
Idea
Is it possible to use additional information to decide which direction to search in?
Yes!
Instead of searching in all directions, let's bias search in the direction of the goal.
![Page 8: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/8.jpg)
Example
Stright-line distances to Bucharest
![Page 9: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/9.jpg)
Example
Start state
Goal state
Expand states in order of their distance to the goal
– for each state that you put on the fringe: calculate straight-line distance to the goal
– expand the state on the fringe closest to the goal
![Page 10: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/10.jpg)
Example
Start state
Goal state
Expand states in order of their distance to the goal
– for each state that you put on the fringe: calculate straight-line distance to the goal
– expand the state on the fringe closest to the goal
Heuristic:
Greedy search
![Page 11: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/11.jpg)
Greedy Search
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 12: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/12.jpg)
Greedy Search
Each time you expand a state, calculate the heuristic for each of the states that you add to the fringe.
– heuristic:
– on each step, choose to expand the state with the lowest heuristic value.
i.e. distance to Bucharest
![Page 13: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/13.jpg)
Greedy Search
Each time you expand a state, calculate the heuristic for each of the states that you add to the fringe.
– heuristic:
– on each step, choose to expand the state with the lowest heuristic value.
i.e. distance to Bucharest
This is like a guess about how far the state is from the goal
![Page 14: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/14.jpg)
Example: Greedy Search
![Page 15: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/15.jpg)
Example: Greedy Search
![Page 16: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/16.jpg)
Example: Greedy Search
![Page 17: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/17.jpg)
Example: Greedy Search
Path: A-S-F-B
![Page 18: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/18.jpg)
Example: Greedy Search
Notice that this is not the optimal path!
Path: A-S-F-B
![Page 19: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/19.jpg)
Example: Greedy Search
Notice that this is not the optimal path!
Path: A-S-F-B
Greedy Search:– Not optimal– Not complete– But, it can be very fast
![Page 20: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/20.jpg)
Greedy vs UCS
Greedy Search:– Not optimal– Not complete– But, it can be very fast
UCS:– Optimal– Complete– Usually very slow
![Page 21: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/21.jpg)
Greedy vs UCS
Greedy Search:– Not optimal– Not complete– But, it can be very fast
UCS:– Optimal– Complete– Usually very slow
Can we combine greedy and UCS???
![Page 22: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/22.jpg)
Greedy vs UCS
Greedy Search:– Not optimal– Not complete– But, it can be very fast
UCS:– Optimal– Complete– Usually very slow
Can we combine greedy and UCS???
YES: A*
![Page 23: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/23.jpg)
Greedy vs UCS
UCS
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 24: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/24.jpg)
Greedy vs UCS
UCS Greedy
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 25: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/25.jpg)
Greedy vs UCS
UCS Greedy
A*
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 26: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/26.jpg)
A*
Image: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 27: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/27.jpg)
A*
: a state
: minimum cost from start to
: heuristic at (i.e. an estimate of remainingcost-to-go)
UCS: expand states in order of
Greedy: expand states in order of
A*: expand states in order of
![Page 28: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/28.jpg)
A*
: a state
: minimum cost from start to
: heuristic at (i.e. an estimate of remainingcost-to-go)
UCS: expand states in order of
Greedy: expand states in order of
A*: expand states in order of
What is “cost-to-go”?
![Page 29: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/29.jpg)
A*
: a state
: minimum cost from start to
: heuristic at (i.e. an estimate of remainingcost-to-go)
UCS: expand states in order of
Greedy: expand states in order of
A*: expand states in order of
What is “cost-to-go”?– minimum cost required
to reach a goal state
![Page 30: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/30.jpg)
A*
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
S a d
b
Gh=5
h=6
h=2
1
8
11
2h=6
h=0
ch=7
3
e h=1
1
S
a
b
c
ed
dG
G
g = 0 h=6
g = 1 h=5
g = 2 h=6
g = 3 h=7
g = 4 h=2
g = 6 h=0
g = 9 h=1
g = 10 h=2
g = 12 h=0
A*: expand states in order of
![Page 31: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/31.jpg)
When should A* terminate?
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
S
B
A
G
2
3
2
2h = 1
h = 2
h = 0h = 3
Should we stop when we enqueue a goal?
No: only stop when we dequeue a goal
![Page 32: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/32.jpg)
Is A* optimal?
A
GS
1 3
h = 6
h = 0
5
h = 7
What went wrong?Actual cost-to-go < heuristicThe heuristic must be less than the actual cost-to-go!
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 33: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/33.jpg)
When is A* optimal?
It depends on whether we are using the tree search or the graph search version of the algorithm.
Recall: – in tree search, we do not track the explored set– in graph search, we do
![Page 34: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/34.jpg)
Recall: Breadth first search (BFS)
What is the purpose of the explored set?
![Page 35: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/35.jpg)
When is A* optimal?
It depends on whether we are using the tree search or the graph search version of the algorithm.
Optimal if h is admissibleOptimal if h is consistent
![Page 36: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/36.jpg)
When is A* optimal?
It depends on whether we are using the tree search or the graph search version of the algorithm.
Optimal if h is admissible
– h(s) is an underestimate of the true cost-to-go.
Optimal if h is consistent
– h(s) is an underestimate of the cost of each arc.
![Page 37: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/37.jpg)
When is A* optimal?
It depends on whether we are using the tree search or the graph search version of the algorithm.
Optimal if h is admissible
– h(s) is an underestimate of the true cost-to-go.
Optimal if h is consistent
– h(s) is an underestimate of the cost of each arc.
What is “cost-to-go”?– minimum cost required
to reach a goal state
![Page 38: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/38.jpg)
When is A* optimal?
It depends on whether we are using the tree search or the graph search version of the algorithm.
Optimal if h is admissible
– h(s) is an underestimate of the true cost-to-go.
Optimal if h is consistent
– h(s) is an underestimate of the cost of each arc.
More on this later...
![Page 39: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/39.jpg)
Admissibility: Example
Stright-line distances to Bucharest
h(s) = straight-line distance to goal state (Bucharest)
![Page 40: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/40.jpg)
Admissibility
Stright-line distances to Bucharest
h(s) = straight-line distance to goal state (Bucharest)
Is this heuristic admissible???
![Page 41: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/41.jpg)
Admissibility
Stright-line distances to Bucharest
h(s) = straight-line distance to goal state (Bucharest)
Is this heuristic admissible???YES! Why?
![Page 42: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/42.jpg)
Admissibility: Example
h(s) = ?
Start state Goal state
Can you think of an admissible heuristic for this problem?
![Page 43: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/43.jpg)
Admissibility
Why isn't this heuristic admissible?
A
GS
1 3
h = 6
h = 0
5
h = 7
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 44: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/44.jpg)
Consistency
What went wrong?
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
S
A
B
C
G
1
1
1
23
h=2
h=1
h=4h=1
h=0
S (0+2)
A (1+4) B (1+1)
C (2+1)
G (5+0)
C (3+1)
G (6+0)
State space graph Search tree
![Page 45: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/45.jpg)
Consistency
Cost of going from s to s'
s s'
![Page 46: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/46.jpg)
Consistency
Rearrange terms
![Page 47: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/47.jpg)
Consistency
Cost of going from s to s' implied by heuristic
Actual cost of going from s to s'
![Page 48: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/48.jpg)
Consistency
Cost of going from s to s' implied by heuristic
Actual cost of going from s to s'
![Page 49: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/49.jpg)
Consistency
Consistency implies that the “f-cost” never decreases along any path to a goal state.– the optimal path gives a goal state its lowest f-cost.
A* expands states in order of their f-cost.
Given any goal state, A* expands states that reach the goal state optimally before expanding states the reach the goal state suboptimally.
![Page 50: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/50.jpg)
Consistency implies admissibility
Suppose:
Then:
![Page 51: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/51.jpg)
Consistency implies admissibility
Suppose:
Then:
![Page 52: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/52.jpg)
Consistency implies admissibility
Suppose:
Then: admissible
![Page 53: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/53.jpg)
Consistency implies admissibility
Suppose:
Then:
![Page 54: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/54.jpg)
Consistency implies admissibility
Suppose:
Then:
admissible
![Page 55: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/55.jpg)
Consistency implies admissibility
Suppose:
Then:
admissibleadmissible
![Page 56: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/56.jpg)
Consistency implies admissibility
Suppose:
Then:
![Page 57: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/57.jpg)
A* vs UCS
Uniform-cost expands equally in all “directions”
A* expands mainly toward the goal, but does hedge its bets to ensure optimality
Start Goal
Start Goal
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
![Page 58: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/58.jpg)
A* vs UCS
Slide: Adapted from Berkeley CS188 course notes (downloaded Summer 2015)
Greedy UCS A*
![Page 59: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/59.jpg)
Choosing a heuristic
The right heuristic is often problem-specific.
But it is very important to select a good heuristic!
![Page 60: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/60.jpg)
Choosing a heuristic
How much better is ?
Consider the 8-puzzle:
: number of misplaced tiles
: sum of manhattan distances between each tile and its goal.
![Page 61: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/61.jpg)
Choosing a heuristic
Consider the 8-puzzle:
: number of misplaced tiles
: sum of manhattan distances between each tile and its goal.
Average # states expanded on a random depth-24 puzzle:
(by depth 12)
![Page 62: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/62.jpg)
Choosing a heuristic
Consider the 8-puzzle:
: number of misplaced tiles
: sum of manhattan distances between each tile and its goal.
Average # states expanded on a random depth-24 puzzle:
(by depth 12)
So, getting the heuristic right can speed things up by multiple orders of magnitude!
![Page 63: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/63.jpg)
Choosing a heuristic
Consider the 8-puzzle:
: number of misplaced tiles
: sum of manhattan distances between each tile and its goal.
Why not use the actual cost to goal as a heuristic?
![Page 64: Heuristic Search - College of Computer and Information … Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA](https://reader031.fdocuments.net/reader031/viewer/2022022500/5aa1e13c7f8b9a84398c4886/html5/thumbnails/64.jpg)
How to choose a heuristic?
Nobody has an answer that always works.
A couple of best-practices:– solve a relaxed version of the problem– solve a subproblem