Search

1
Search

• Exhaustive/Blind Search Methods
• Depth First Search
• Breadth First Search
• Heuristic Methods
• Hill Climbing
• Beam Search
• Best First Search

2
Goal

• You should be able to tell if search is the best
  way to solve a problem and
• If so, which search method is most suitable

3
Find a Path
4
A
B
C
4
3
5
5
s
G
4
3
F
D
E
2
4
4
Find a Path

• Shortest Path
• Any Path
• Blind Search
• BFS
• DFS

5
Search Tree

• List all possible paths
• Eliminate cycles from paths
• Result A search tree
• A special kind of semantic tree where each node
  denotes a path

6
Search Tree Terminology

• Root Node
• Leaf Node
• Ancestor/Descendant
• Branching factor
• Complete path/Partial path
• Expanding open nodes results in closed nodes

7
Search Trees - Analysis

• Exponential explosion of possible paths
• Branching factor b, Depth d, how many possible
  paths?

8
Depth First Search (DFS)

• Form a one element queue consisting of a
  zero-length path that only contains root
• Until the first path in the queue terminates at
  the goal node or the queue is empty
• Remove the first path from the queue create new
  paths by extending the first path to all
  neighbors of the terminal node
• Reject all new paths with loops
• Add the new paths if any to the FRONT of the
  queue
• If the goal node is found?success else failure

9
Breadth First Search (BFS)

• Form a one element queue consisting of a
  zero-length path that only contains root
• Until the first path in the queue terminates at
  the goal node or the queue is empty
• Remove the first path from the queue create new
  paths by extending the first path to all
  neighbors of the terminal node
• Reject all new paths with loops
• Add the new paths if any to the BACK of the queue
• If the goal node is found?success else failure

10
BFS or DFS

• DFS is good when partial paths quickly reach dead
  ends or become complete paths. DFS is bad when
  there are long paths (infinite)
• BFS works on trees that are effectively
  infinitely deep, bad when all paths lead to the
  goal node at more or less the same depth. BFS is
  not good when b is large

11
NonDeterministic Search

• Form a one element queue consisting of a
  zero-length path that only contains root
• Until the first path in the queue terminates at
  the goal node or the queue is empty
• Remove the first path from the queue create new
  paths by extending the first path to all
  neighbors of the terminal node
• Reject all new paths with loops
• Add the new paths at RANDOM places in the queue
• If the goal node is found?success else failure

12
NDFS

• When you do not have enough information to choose
  between BFS and DFS