A Search

$$f(n)=g(n)+h(n)$$

Where $g(n)$ is the cost from the start node to node n Where $h(n)$ heuristic estimate from n to a goal

• maintains a priority queue ordered by $f(n)$
• expands the node with the smallest $f(n)$
• combines search depth (g) and goal guidance (h)

this is a framework, not an algorithm.

if $h(n)$ overestimates the true remaining cost, algorithm A may expand a suboptimal path first