Problem in AI
The question that has to be answered is the problem. A problem must be precisely specified to be solved. Determining the start state, goal state, various admissible states, and transitions are all included in the definition.
Hence, an issue is made up of a set of path restrictions, a set of goal states, and a set of initial states. Finding a path that ends at any objective state while satisfying the path limitations is the challenge.
Four elements are defined in a fairly formal approach to formulate any problem.
These four elements consist
- The first or initial condition.
- All conceivable states that can be reached from the starting state are described by state space.
- A goal test that establishes the goal status of a given condition.
- Path cost is a function that gives every path a numerical cost.
Problem Characteristics in AI
The following characteristics are used in the analysis of a problem:
1. Can the issue be broken down?
2. Are steps in a solution reversible or ignorable?
3. Is the universe with problems predictable?
4. A decent solution is it relative or absolute?
5. Is a state or a route to a state the answer?
6. What function does knowledge serve?
7. Does the task involve interacting with people?
These features are examined in AI tasks like Tower of Hanoi, Traveling Salesman, Water Jug, Chess, and 8-Puzzle.
Example of Problem Characteristics:
Problem Characteristics |
Satisfied |
1. Is the problem decomposable? |
No, because one game has a single solution. |
2. Can solution steps be ignored or undone? |
No, as in the actual game steps can't undo. |
3. Is the problem universe predictable> |
No, since the move of another player is not sure. |
4. Is a good solution absolute or relative? |
Absolute |
5. Is the solution a state or a path to state? |
Path because winning state describes the path to state. |
6. What is the role of knowledge? |
It helps to constrain solution search. |
7. Does the task require human interactions. |
No, since additional assistance is not required. |