• Completeness: Will the algorithm eventually find a solution if one exists?

    • depends on branching factor, cycles, and termination conditions
  • Optimality: Will the algorithm return the best solution according to the cost function?

    • requires admissible evaluation or exhaustive guarantees
    • often sacrificed in exchange for speed or scalability
  • Time complexity: How many nodes are generated/expanded?

    • Typically exponential in depth for uninformed search
    • Strongly affected by branching factor and heuristic guidance
  • Space complexity: How much memory is retained during search?

    • Includes frontier/fringe storage and closed lists
    • In practice, this is often the first limiting factor
    • Motivates memory-bounded and local search methods