Steepest Hill climbing/Best first search/A* Algorithm

Steepest Hill climbing differs from simple hill climbing by choosing the Best Successor rather than the first one.This indicates that it has elements of breadth first  search algorithm

The Algorithm in a simplified manner is here 

1.Evaluate the initial State.

2.If it is goal state than Quit otherwise make Initial state as the current state and proceed.

3 Repeat  

   set target to be the State that any Successor of the Current state can be better.

4.For (in a loop)  Each operator that can be applied in the Current State Apply new operator to new       state.if it is a goal state than Quit otherwise Compare with the Target state if better than set Current state to target state.

Best First Search   

We had BFS and DFS in DFS solution can be found without Computing the nodes and in BFS is good because it does not get trapped in dead ends.

But for Best first search it takes the benefit of both the steps (DFS and BFS) takes the most step to be chosen . For example if one of the nodes chosen generates nodes that are less promising it is possible to choose the same level other node (So same level is the word means we can assume it to be BFS)So it means it has switched form DFS to BFS.

A* Algorithm

A* algorithm combines the features of uniform cost search and + pure heuristic to efficiently compute the optimal search.A* is the best First search algorithm in which cost  associated with the Node is f(n)=g(n)+h(n).

where g(n)⟶ cost from initial node to node n.

where h(n)⟶heuristic estimate or cost or path from node n to goal.

f(n) estimates the lowest solution path going through node n .At each node choose the lowest f to expand.And if modes have the same f value than choose the least h value that is the heuristic cost estimated to reach the goal.The algorithm terminates when the goal is chosen for expansion.