UF CAP 4621 Artificial Intelligence HW3Solution

From University
Jump to: navigation, search

The graph shown below represents a network of cities (labeled A through J) with distances between each pair of cities given as the value on each arc. A table is given for each of the nodes identifying its heuristic estimated distance from the goal. You are to determine the shortest path between A and J. You can drop nodes from the tree that you grow that are elsewhere in the tree with a lower cost. NOTE: For your answers you should draw a SERIES of partial solution trees showing the expansion of the search space. DO NOT DRAW YOUR ANSWER AS A SINGLE TREE!

Note: I think all of these trees are correct. I might have missed a few pruning cases when I constructed the diagrams.

A. Find a solution using Uniform cost search.
CAP4621HW3uc1.jpg
CAP4621HW3uc2.jpg
CAP4621HW3uc3.jpg
CAP4621HW3uc4.jpg
CAP4621HW3uc5.jpg
CAP4621HW3uc6.jpg
CAP4621HW3uc7.jpg
CAP4621HW3uc8.jpg

B. Find a solution using Best-First Search.
CAP4621HW3bf1.jpg
CAP4621HW3bf2.jpg
CAP4621HW3bf3.jpg
CAP4621HW3bf4.jpg

C. Use the A* algorithm to determine the shortest path.
CAP4621HW3astar1.jpg
CAP4621HW3astar2.jpg
CAP4621HW3astar3.jpg
CAP4621HW3astar4.jpg
CAP4621HW3astar5.jpg
CAP4621HW3astar6.jpg
CAP4621HW3astar7.jpg


D. Is the solution found by A* the best solution? Why or why not?

Yes it is but just by chance. There is one node, C, which is not admissible!

Internal Links

Parent Article: UF CAP 4621 Artificial Intelligence Homework